United States
Department
of Agriculture
Forest Service
Rocky Mountain
Research Station
Proceedings
RMRS-P-31
February 2004
Seed and Soil Dynamics in
Shrubland Ecosystems:
Proceedings
Abstract
Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance; McArthur, E. Durant, comps. 2004. Seed and soil dynamics
in shrubland ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proc. RMRS-P-31. Fort Collins, CO: U.S.
Department of Agriculture, Forest Service, Rocky Mountain Research Station. 216 p.
The 38 papers in this proceedings are divided into six sections; the first includes an overview paper and documentation of the first
Shrub Research Consortium Distinguished Service Award. The next four sections cluster papers on restoration and revegetation,
soil and microsite requirements, germination and establishment of desired species, and community ecology of shrubland systems.
The final section contains descriptions of the field trips to the High Plains Grassland Research Station and to the Snowy Range and
Medicine Bow Peak. The proceedings unites many papers on germination of native seed with vegetation ecology, soil physiochemical properties, and soil biology to create a volume describing the interactions of seeds and soils in arid and semiarid shrubland
ecosystems.
Keywords: wildland shrubs, seed, soil, restoration, rehabilitation, seed bank, seed germination, biological soil crusts
Acknowledgments
The symposium, field trips, and subsequent publication of this volume were made possible through the hard work of
many people. We wish to thank everyone who took a part in ensuring the success of the meetings, trade show, and paper
submissions. We thank the University of Wyoming Office of Academic Affairs, the Graduate School, and its Dean, Dr.
Steve Williams, for their assistance. We recognize the Student Union Staff and the University of Wyoming Conferences
and Institute Staff for logistic support. Especially helpful in coordinating the meetings were Kimm Malody, Kelli Belden,
and Cindy Wood. Jennifer Muscha (Boyle), Leah Burgess, Courtney Ladenburger, Brian Mealor, and Sonja Parson of
the Department of Renewable Resources provided excellent help during the sessions. We also thank all of the trade
show contributors, especially Truax, Inc., for their generous support. Technical sessions were chaired by D. Terrance
Booth, Leah Burgess, Kathleen Dwire, C. Lynn Kinter, Susan Meyer, Brian Mealor, Steve Monsen, Jean D. Reeder,
Bruce Roundy, and Steve Williams. Mary I. Williams provided original artwork for the cover. We thank the Rocky
Mountain Research Station Publishing Services and Louise Kingsbury for preparing the final publication.
Ann L. Hild
Nancy L. Shaw
Susan E. Meyer
D. Terrance Booth
E. Durant McArthur
Compilers
Publisher’s note: Papers in this report were reviewed by the compilers. Rocky Mountain Research Station Publishing
Services reviewed papers for format and style. Authors are responsible for content.
Illustration on the cover is by Mary I. Williams. 2002. Drawing of winterfat. Used with permission.
Seed and Soil Dynamics in
Shrubland Ecosystems:
Proceedings
Laramie, WY, August 12–16, 2002
Compilers
Ann L. Hild, Department of Renewable Resources, University of Wyoming, Laramie, WY.
Nancy L. Shaw, U.S. Department of Agriculture, Forest Service, Rocky Mountain
Research Station, Boise, ID.
Susan E. Meyer, U.S. Department of Agriculture, Forest Service, Rocky Mountain
Research Station, Shrub Sciences Laboratory, Provo, UT.
D. Terrance Booth, U.S. Department of Agriculture, Agricultural Research Service,
High Plains Grassland Research Station, Cheyenne, WY.
E. Durant McArthur, Project Leader, U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station, Shrub Sciences Laboratory, Provo, UT.
Shrub Research Consortium
USDA Forest Service, Rocky Mountain Research Station, Shrub Sciences Laboratory*, Provo, Utah, E. Durant
McArthur (chair)
Brigham Young University*, Provo, Utah, Daniel J. Fairbanks
USDA Agricultural Research Service, Renewable Resource Center*, Reno, Nevada, James A. Young
Utah State University*, Logan, Eugene W. Schupp
State of Utah, Department of Natural Resources, Division of Wildlife Resources*, Ephraim, Scott C. Walker
University of California, Los Angeles, Philip W. Rundel
Colorado State University, Fort Collins, William K. Lauenroth
University of Idaho, Moscow, Steven J. Brunsfeld
University of Montana, Missoula, Donald J. Bedunah
Montana State University, Bozeman, Carl L. Wambolt
University of Nevada, Reno, Paul T. Tueller
University of Nevada, Las Vegas, Stanley D. Smith
Oregon State University, Corvallis, Lee E. Eddleman
New Mexico State University, Las Cruces, Kelly W. Allred
Texas A & M System, Texas Agricultural Experiment Station, San Angelo, Darrell N. Ueckert
Texas Tech University, Lubbock, Ronald E. Sosebee
USDA Agricultural Research Service, High Plains Grassland Research Station, Cheyenne, Wyoming, D. Terrance
Booth
USDA Agricultural Research Service, Jornada Experimental Range, Las Cruces, New Mexico, Jerry R. Barrow
USDA Forest Service, Rocky Mountain Research Station, Great Basin Ecology Laboratory, Reno, Nevada, Robin J.
Tausch
Universtiy of Utah, Salt Lake City, James R. Ehleringer
Weber State University, Ogden, Utah, Barbara A. Wachocki
Battelle Pacific Northwest Laboratories, Richland, Washington, Janelle L. Downs
Bechtel Nevada, Las Vegas, W. Kent Ostler
University of Wyoming, Laramie, Ann L. Hild
*Charter members
The Shrub Research Consortium (SRC) was formed in 1983 with five charter members (see list). Over time SRC has
grown to its present size of 25 institutional members. The SRC had three principal objectives in its charter: (1) developing
plant materials for shrubland rehabilitation; (2) developing methods of establishing, renewing, and managing
shrublands in natural settings; and (3) assisting with publication and dissemination of research results. These objectives
have been met by a series of symposia sponsored by the Consortium and partners. This publication is the 12th in the
series; the previous 11 are listed on the next page. The U.S. Department of Agriculture, Forest Service, Intermountain
Research Station and Rocky Mountain Research Station have published proceedings of all symposia to date. The
executive committee has plans for another symposium in 2004 in Lubbock, Texas, with the theme of Shrublands Under
Fire. Each symposium has had a theme, but the executive committee has encouraged attendance and participation by
shrubland ecosystem biologists and managers with wider interests than any particular symposium theme—the heart
of the Consortium’s programs are wildland shrub ecosystem biology, research, and management.
iii
Availability of Previous Wildland Shrub Symposia Proceedings
First:
Tiedemann, A. R.; Johnson, K. L., compilers. 1983. Proceedings—research and management of bitterbrush and cliffrose in Western North America; 1982 April 13–15; Salt Lake City, UT. General Technical
Report INT-152. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest and
Range Experiment Station. 279 p. Out of print—available from National Technical Information Service as
document PB83-261537 A13.
Second:
Tiedemann, A. R.; McArthur, E. D.; Stutz, H. C.; Stevens, R.; Johnson, K. L., compilers. 1984. Proceedings— symposium on the biology of Atriplex and related chenopods; 1983 May 2–6; Provo, UT. General
Technical Report INT-172. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station. 309 p. Out of print—available from National Technical Information
Service as document PB85-116358 A14.
Third:
McArthur, E. D.; Welch, B. L., compilers. 1986. Proceedings—symposium on the biology and management
of Artemisia and Chrysothamnus; 1984 July 9–13; Provo, UT. General Technical Report INT-200. Ogden,
UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station. 398 p. Out of print—
available from National Technical Information Service as document PB86-182318 A18.
Fourth:
Provenza, F. D.; Flinders, J. T.; McArthur, E. D., compilers. 1987. Proceedings—symposium on plant
herbivore interactions; 1985 August 7–9; Snowbird, UT. General Technical Report INT-222. Ogden, UT:
U.S. Department of Agriculture, Forest Service, Intermountain Research Station. 179 p. A few copies are
available from the Rocky Mountain Research Station; otherwise available from National Technical
Information Service as document PB 90-228578 A09.
Fifth:
Wallace, A.; McArthur, E. D.; Haferkamp, M. R., compilers. 1989. Proceedings—symposium on shrub
ecophysiology and biotechnology; 1987 June 30–July 2; Logan, UT. General Technical Report INT-256.
Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station. 183 p. Out
of print—available from National Technical Information Service as document PB89-156442 A09
Sixth:
McArthur, E. D.; Romney, E. M.; Smith, S. D.; Tueller, P. T., compilers. 1990. Proceedings—symposium
on cheatgrass invasion, shrub die-off, and other aspects of shrub biology and management; 1989 April
5–7; Las Vegas, NV. General Technical Report INT-276. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Research Station. 351 p. Out of print—available from National Technical
Information Service as document PB91-117275 A16.
Seventh:
Clary, W. P.; McArthur, E. D.; Bedunah, D.; Wambolt, C. L., compilers. 1992. Proceedings—symposium
on ecology and management of riparian shrub communities; 1991 May 29–31; Sun Valley, ID. General
Technical Report INT-289. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain
Research Station. 232 p. Out of print—available from National Technical Information Service as document
PB92-227784 A11.
Eighth:
Roundy, B. A.; McArthur, E. D.; Haley, J. S.; Mann, D. K., compilers. 1995. Proceedings: wildland shrub
and arid land restoration symposium; 1993 October 19–21; Las Vegas, NV. General Technical Report INTGTR-315. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station.
384 p. Available from Rocky Mountain Research Station.
Ninth:
Barrow, J. R.; McArthur, E. D.; Sosebee, R. E.; Tausch, R. J., compilers. 1996. Proceedings: shrubland
ecosystem dynamics in a changing environment; 1995 May 23–25; Las Cruces, NM. General Technical
Report INT-GTR-338. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station. 275 p. A few copies are available from the Rocky Mountain Research Station; otherwise
available from National Technical Information Service as document PB 96-178637 A09.
Tenth:
McArthur, E. D.; Ostler, W. K.; Wambolt, C. L. compilers. 1999. Proceedings: shrubland ecosystem
ecotones; 1998 August 12–14; Ephraim, UT. Proceedings RMRS-P-11. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Rocky Mountain Research Station. 299 p. Available from Rocky Mountain
Research Station.
Eleventh:
McArthur, E. D.; Fairbanks, D. J., compilers. 2001. Shrubland ecosystem genetics and biodiversity:
proceedings; 2000 June 13–15; Provo, UT. Proc. RMRS-P-21. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Rocky Mountain Research Station. 365 p. Available from Rocky Mountain Research
Station.
iv
Contents
Page
Introduction and Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Nancy L. Shaw
Ann L. Hild
Seed and Soil Dynamics in Shrubland Ecosystems: Introduction . . . . . . . . . . . . . . 3
Ann L. Hild
Nancy L. Shaw
Shrub Research Consortium: Distinguished Service Award
Honoring E. Durant McArthur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Restoration and Revegetation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Edith B. Allen
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus:
A Comparison Between Southern California and the Intermountain Region . . . . . . 9
C. Lynn Kinter
Richard N. Mack
Comparing Phenotype and Fitness of Native, Naturalized, and Invasive
Populations of Downy Brome (Cheatgrass, Bromus tectorum) . . . . . . . . . . . . . . . 18
Mary I. Williams
Mohammed Tahboub
John T. Harrington
David R. Dreesen
April Ulery
Root Growth of Apache Plume and Serviceberry on Molybdenum Mine
Overburden in Northern New Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Kathleen J. Miglia
D. Carl Freeman
E. Durant McArthur
Bruce N. Smith
Importance of Genotype, Soil Type, and Location on the Performance of
Parental and Hybrid Big Sagebrush Reciprocal Transplants in the Gardens of
Salt Creek Canyon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Robert R. Blank
James A. Young
Revegetation of Saline Playa Margins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Laurel E. Vicklund
Gerald E. Schuman
Ann L. Hild
Influence of Sagebrush and Grass Seeding Rates on Sagebrush
Density and Plant Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Soil Components and Microsites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Thomas A. Monaco
Jay B. Norton
Douglas A. Johnson
Thomas A. Jones
Jeanette M. Norton
Soil Nitrogen Controls on Grass Seedling Tiller Recruitment . . . . . . . . . . . . . . . . . 47
Robert R. Blank
Enzyme Activity in Temperate Desert Soils: Influence of Microsite,
Depth, and Grazing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Jerry R. Barrow
Unique Characteristics of a Systemic Fungal Endophyte of Native
Grasses in Arid Southwestern Rangelands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Jay B. Norton
Thomas A. Monaco
Jeanette M. Norton
Douglas A. Johnson
Thomas A. Jones
Cheatgrass Invasion Alters Soil Morphology and Organic Matter
Dynamics in Big Sagebrush-Steppe Rangelands . . . . . . . . . . . . . . . . . . . . . . . . . . 57
R. L. Pendleton
B. K. Pendleton
G. L. Howard
S. D. Warren
Response of Lewis Flax Seedlings to Inoculation With
Arbuscular Mycorrhizal Fungi and Cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . . 64
v
J. D. Reeder
D. W. Grant
G. G. Jezile
R. D. Child
Heterogeneity in Soil C and N in Late- and Mid-seral
Sagebrush-Grass Rangeland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Roger Rosentreter
David J. Eldridge
Monitoring Rangeland Health: Using a Biological
Soil Crust Stability Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
David J. Eldridge
Roger Rosentreter
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community in
Southwestern Idaho, U.S.A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Stephen E. Williams
Joe D. Johnson
Larry C. Munn
Terry Nieder
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by
Cercocarpus montanus Raf. and Purshia tridentata (Pursh) DC. . . . . . . . . . . . . . . 84
Germination and Seedling Establishment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
D. T. Booth
Y. Bai
E. E. Roos
Cultural Methods for Enhancing Wyoming Big Sagebrush
Seed Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Ann DeBolt
Carol S. Spurrier
Seeds of Success and the Millennium Seed Bank Project . . . . . . . . . . . . . . . . . . 100
Susan C. Garvin
Susan E. Meyer
David L. Nelson
Recruitment Failure of Chenopod Shrub Populations
in the Great Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
M. R. Haferkamp
M. D. MacNeil
Annual Brome Seed Germination in the Northern Great Plains: An Update . . . . 115
Robert Hammon
Gary Noller
Fate of Fall-Sown Bitterbrush Seed at Maybell, Colorado . . . . . . . . . . . . . . . . . . 120
Susan E. Meyer
Stephanie L. Carlson
Comparative Seed Germination Biology and Seed Propagation
of Eight Intermountain Species of Indian Paintbrush . . . . . . . . . . . . . . . . . . . . . . 125
R. L. Pendleton
B. K. Pendleton
G. L. Howard
S. D. Warren
Effects of Biological Soil Crusts on Seedling Growth and
Mineral Content of Four Semiarid Herbaceous Plant Species . . . . . . . . . . . . . . . 131
Michael P. Schellenberg
Seeding Time and Method for Winterfat in the Semiarid Region
of the Canadian Prairies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
L. St. John
P. Blaker
New Native Plant Releases from the USDA-NRCS,
Aberdeen, ID, Plant Materials Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Jeffrey R. Taylor
Bruce A. Roundy
Phil S. Allen
Susan E. Meyer
Predicting Seedling Emergence Using Soil Moisture
and Temperature Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Community Ecology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Donald J. Bedunah
Kristi D. Pflug
Thad E. Jones
Antelope Bitterbrush Flowering, Survival, and Regeneration Following
Ponderosa Pine Forest Restoration Treatments . . . . . . . . . . . . . . . . . . . . . . . . . 149
Justin D. Derner
Charles R. Tischler
H. Wayne Polley
Hyrum B. Johnson
Effects of Elevated CO2 on Growth Responses of Honey Mesquite Seedlings
From Sites Along a Precipitation Gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
vi
Charles R. Tischler
Justin D. Derner
H. Wayne Polley
Hyrum B. Johnson
Responses of Seedlings of Five Woody Species to
Carbon Dioxide Enrichment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Simon A. Lei
Soil Moisture Attributes of Three Inland Sand Dunes
in the Mojave Desert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Simon A. Lei
Leaf Temperatures of Blackbrush: Relationships With Air
and Soil Surface Temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Jennifer M. Muscha
Ann L. Hild
Larry C. Munn
Pete D. Stahl
Impacts of Livestock Exclusion From Wyoming Big
Sagebrush Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
W. A. Laycock
D. L. Bartos
K. D. Klement
Species Richness Inside and Outside Long-Term Exclosures . . . . . . . . . . . . . . . 183
J. Kent McAdoo
Sherman R. Swanson
Brad W. Schultz
Peter F. Brussard
Vegetation Management for Sagebrush-Associated Wildlife Species . . . . . . . . . 189
Carl L. Wambolt
Trista Hoffman
Browsing Effects on Wyoming Big Sagebrush
Plants and Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
Kristene A. Partlow
Richard A. Olson
Gerald E. Schuman
Scott E. Belden
Effects of Wildlife Utilization and Grass Seeding Rates on Big Sagebrush
Growth and Survival on Reclaimed Mined Lands . . . . . . . . . . . . . . . . . . . . . . . . . 198
Field Trips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
D. T. Booth
J. D. Derner
Research at the High Plains Grassland Research Station . . . . . . . . . . . . . . . . . . 209
Larry Munn
C. Lynn Kinter
Soils and Vegetation of the Snowy Range, Southeastern Wyoming
Wildland Shrub Symposium Field Trip: Laramie to Woods Landing,
Riverside, Libby Flats, and Centennial, Wednesday, August 14, 2002 . . . . . . . . 212
vii
Introduction and
Overview
Restoration and
Revegetation
Soil Components and
Microsites
Germination
and Seedling
Establishment
Community
Ecology
Field Trips
Introduction and
Overview
1
Illustration on the reverse side is by Eva Teseo and Meggan Laxalt. In: Engwistle, P. G.; DeBolt, A. M.; Kaltenecker, J. H.; Steenhof, K., comps. 2000.
Proceedings: sagebrush steppe ecosystems symposium. Pub. BLM/ID/PT-001001+1150. Boise, ID: U.S. Department of the Interior, Bureau of Land
Management. 145 p.
Seed and Soil Dynamics in
Shrubland Ecosystems:
Introduction
Nancy L. Shaw
Ann L. Hild
This proceedings is the twelfth in a series on the biology
and management of wildland shrub ecosystems, sponsored
by the Shrub Research Consortium and published by the
Rocky Mountain Research Station. Other cosponsors of the
symposium on Seed and Soil Dynamics in Shrubland Ecosystems included the University of Wyoming Department of
Renewable Resources and the USDA-ARS High Plains Grassland Research Station.
The twelfth wildland shrub symposium united above- and
below-ground research on shrublands to more clearly delineate the dynamic nature of shrubland ecosystems. The
keynote address (Allen, this proceedings) characterized differences in exotic brome invasions and impacts depending
on geographic locations and poses some interesting questions relative to the common trends of invasion of annual
grasses. The meeting also provided the venue for presentation of the first Shrub Research Consortium Distinguished
Service Award to Dr. E. Durant McArthur, Project Leader,
Shrub Sciences Laboratory, USDA Forest Service, in Provo,
UT (Hild and Shaw, this proceedings).
Our symposium intended to unite plant ecology academics, managers, and soil scientists into a dialogue to consider
below-ground dynamics and their ties to shrubland ecosystem restoration and management. The 38 papers in this
volume report research on below-ground biology, seedbanks,
biological soil crusts, microsite and soil surface ecology,
invasive species competition, restoration and revegetation
in sagebrush steppe, blackbrush, salt desert shrub, annual
grasslands, Chihuahuan and Sonoran Desert shrublands,
and mountain shrub communities. Field trips included travel
from Laramie west through the Snowy Mountains near
Riverside and Encampment to examine vegetative ecology,
soils, and geomorphology of the Snowy Range. Travel to the
east from Laramie included the USDA-Agricultural Research Service High Plains Grassland Research Station in
Cheyenne, WY, where participants examined long-term
research on grazing and carbon sequestration. The poster
session included 42 academic poster presentations and numerous trade exhibitors.
The thirteenth Wildland Shrub Symposium will be held in
Lubbock, TX, in summer 2004, and will consider the theme
“Shrublands Under Fire.” Previous symposia cover a wide
variety of topics relevant to shrubland ecosystems (see front
pages of this proceedings for their availability).
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Nancy L. Shaw is a Research Botanist, USDA Forest Service, Rocky
Mountain Research Station, Boise, ID 83702, U.S.A. Ann L. Hild is an
Associate Professor, Department of Renewable Resources, Box 3354, University of Wyoming, Laramie, WY 82071, U.S.A.; e-mail: annhild@uwyo.edu.
USDA Forest Service Proceedings RMRS-P-31. 2004
3
Shrub Research Consortium:
Distinguished Service Award
Honoring E. Durant McArthur
Ann L. Hild
Nancy L. Shaw
In August 2002, at the Twelfth Wildland Shrub Symposium held in Laramie, WY, the Shrub Research Consortium presented its first Distinguished Service Award to Dr.
E. Durant McArthur, Project Leader at the Shrub Sciences
Laboratory in Provo, UT, for his phenomenal contributions
to our understanding of shrubland ecology and biology. The
award honors the long and productive career of this great
shrubland scientist. We include this award paper in the
proceedings to recognize some of the many contributions
made by Durant over his more than 30 years of service with
the USDA Forest Service as a Research Geneticist and
Shrubland Ecologist.
Professional Background _________
Durant graduated from Dixie College in St. George, UT,
with an A.S. degree in 1963. He received a B.S. degree in
molecular and genetic biology (cum laude) in 1965, an M.S.
degree in molecular and genetic biology with a minor in
botany, and a Ph.D. degree in plant genetics in 1970 from the
University of Utah. His doctoral work examined the cytogenic and evolutionary development of aneuploid-tetraploid
Mimulus glabratus (Scrophulariaceae). In 1971 he completed post-doctoral research on the cytogentetics of domesticated and wild Brassicaceae at the University of Leeds,
England.
In 1972 Durant began his career with the USDA Forest
Service, Intermountain Forest and Range Experiment Station, as a Research Geneticist assigned to the Great Basin
Experimental Range in Ephraim, UT. In 1975 he was assigned to the Shrub Sciences Laboratory in Provo, UT,
where, since 1983, he has served as Project Leader of the
Shrubland Biology and Restoration Research Work Unit of
what is now the Rocky Mountain Research Station. Durant
has also served as an Adjunct Faculty member with the
Department of Botany and Range Science at Brigham Young
University since 1976.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Ann L. Hild is an Associate Professor Department of Renewable Resources,
Box 3354, University of Wyoming, Laramie, WY 82071, U.S.A.; e-mail:
annhild@uwyo.edu. Nancy L. Shaw is a Research Botanist, USDA Forest
Service, Rocky Mountain Research Station, Boise, ID 83702, U.S.A.
4
Formation of the Shrub Research
Consortium ____________________
Dr. McArthur was instrumental in organizing the Shrub
Research Consortium in 1983 and has served as Chair of the
Consortium for the last 20 years. Under his guidance the
Shrub Research Consortium has sponsored 12 symposia to
date, providing a forum for shrubland researchers and managers from around the globe to present research, discuss
management techniques, and ultimately advance our knowledge of shrubland ecosystems. Proceedings of these meetings
have been and continue to be valuable references for shrubland
ecologists and land managers, irrespective of their training
or agency affiliation.
Professional Honors and Research
Contributions ___________________
Durant’s contributions have provided the foundation papers on genetic differentiation in Subgenus Tridentatae of
Artemisia and numerous other shrub genera and species.
His recent work on hybridization of subspecies of Artemisia
tridentata amplifies our understanding of evolutionary
change in shrub populations distributed across variable
environmental gradients. His work involves the collaboration of scientists from around the globe. He has published
more than 350 research papers including numerous journal
articles and several book chapters. Durant has been a key
compiler for the Wildland Shrub Symposia Proceedings (10
of 12 proceedings to date). His research expertise includes
selection and breeding of shrubland species for rangeland
rehabilitation, chromosome studies of Artemisia subgenus
Tridentatae, adaptation, breeding system, and seed production of Atriplex canescens, habitat requirements of desert
tortise, and genetic structure of wildland restoration
plantings. He has been directly involved in development of
cultivar and germplasm releases in cooperation with the
USDA-NRCS Plant Materials Program, the Utah Division
of Wildlife Resources, and State Agricultural Experiment
Stations. Species he has assisted with include “Rincon”
fourwing saltbush, “Immigrant” forage kochia, “Hobble
Creek” mountain big sagebrush and “Timp” Utah sweetvetch.
His enthusiasm and guidance have united innovative research in ecology with genetic variation within wildland
populations to form practical applications that have improved shrubland restoration efforts throughout the Intermountain West.
USDA Forest Service Proceedings RMRS-P-31. 2004
Shrub Research Consortium: Distinguished Service Award Honoring E. Durant McArthur
Durant has been honored with numerous awards over the
years, including the USDA Forest Service Superior Scientist
Award (1990) and Distinguished Scientist Award (1996), the
USDA Forest Service Rocky Mountain Research Station
Eminent Scientist Publication Award (2001), and the Society for Range Management Outstanding Achievement Award
(1992). He has served as Chairman of the Shrub Research
Consortium since 1983, President of the Utah Section of the
Society for Range Management (1987), and President of the
Intermountain Consortium for Arid Lands Research (1991 to
present). In 2002, just prior to receiving the Shrub Research
Consortium Award, Durant received the USDA Forest Service New Century of Service Award for the Rocky Mountain
Research Station. We are pleased that the Shrub Research
Consortium is led by such a nationally honored public
servant and distinguished scientist.
Mentorship Legacy ______________
Durant has been a guiding force and has served an
inspirational role for many young scientists (fig. 1). His
genuine encouragement of new scholars has been demonstrated repeatedly in meetings, on field excursions, and in
his role on many M.S. and Ph.D. graduate committees. He is
commonly involved in field trips for student groups (fig. 2)
and visiting scientists (for which the authors are especially
grateful). Durant has devoted endless hours to interagency
Hild and Shaw
cooperation and is always willing to consult with managers
on specific shrubland management issues.
In his role as research leader, he has directed the work of
a long series of talented scientists at the Shrub Sciences
Laboratory. His model scientific abilities and genuine love
for people has allowed Durant to serve as an inspirational
and energetic leader of some of the premier shrubland
researchers in the world.
Personal Accomplishments _______
Durant married Virginia in December 20, 1963, and became the proud father of two sons, Ted and Curt, and two
daughters, Monica and Denise. He is now a proud grandfather of 11 grandchildren (thus far). He has remained active
in many community groups and activities in Provo.
Conclusion _____________________
The Shrub Research Consortium is proud to have the
association and leadership of such a well-respected leader
and well-recognized scientific mind as that of Dr. E. Durant
McArthur. It is with great pride that we presented Durant
with the first Shrub Research Consortium Distinguished
Service Award in gratitude for his tireless efforts in promoting the recognition and understanding of global shrubland
ecosystems.
SHRUB RESEARCH CONSORTIUM
Distinguished Service Award
Presented to
E. DURANT MCARTHUR
For his sustained leadership of the Shrub Research Consortium, his
exemplary research career, and his dedication to enhancing our
understanding of the biology and management of shrublands in
North America and around the globe.
University of Wyoming
Laramie, Wyoming
USDA Forest Service Proceedings RMRS-P-31. 2004
August 13, 2002
5
Hild and Shaw
Shrub Research Consortium: Distinguished Service Award Honoring E. Durant McArthur
Figure 1—USDA Forest Service Intermountain Forest and Range Experiment Station personnel
gathered at the Desert Range Experiment Station on the occassion of a range science retreat, June 8,
1983. Front row (left to right): Mike Stanley, Susan Koniak, Bruce Welch, Renee O’Brien, Ralph
Holmgren, John Kinney. Row 2: Jeanne Chambers, Dale Bartos, Walt Mueggler, Warren Clary.
Row 3: Bob Ferguson, Fred Wagstaff, Sherel Goodrich, Duane Lloyd. Row 4: Art Tiedemann,
Durant McArthur, Keith Evans, Roy Harniss, Rich Everett.
Figure 2—An enthusiastic Durant sharing his research on the University of Wyoming shrubland
ecology graduate class field trip to central Utah in May 2002.
6
USDA Forest Service Proceedings RMRS-P-31. 2004
Restoration and
Revegetation
Aquilegia laramiensis
7
Illustration on the reverse side is by Isobel Nichols. In: Fertig, W.; Refsdal, C.; Whipple, J. 1994. Wyoming rare plant field guide. Cheyenne: Wyoming
Rare Plant Technical Committee. Available at: http://www.npwrc.usgs.gov/resource/distr/others/wyplant/species.htm (Accessed 12/18/03).
Restoration of Artemisia
Shrublands Invaded by Exotic
Annual Bromus: A Comparison
Between Southern California and
the Intermountain Region
Edith B. Allen
Abstract: Sagebrush shrublands of southern California and the
Great Basin are both undergoing massive vegetation-type conversion to exotic annual grasslands dominated by brome species. In
southern California the main players are California sagebrush and
red brome, while the Intermountain Region has big sagebrush and
cheatgrass. Exotic annuals tend to be more dominant in soils with
moderate to high levels of soil P and N. The type conversion of
California sagebrush shrublands has been occurring during the last
25 to 30 years, compared to the last 100 years in the Great Basin.
One reason for this might be the recent increase in anthropogenic N
deposition caused by automobile emissions, up to 30 kg N per ha
year. Field fertilization experiments in southern California also
showed the exotic annual grasses have increased biomass with
elevated N. The increased grass biomass has likely fueled more
frequent fires in California. Even though California shrublands are
adapted to fire compared to the big sagebrush-grasslands, frequent
fire inhibits the re-establishment of shrubs when brome and other
annual grasses are abundant. Restoration attempts have included
timed fire to deplete the annual seedbank, herbicides, mowing,
grazing, and reseeding and replanting. A comparison of some of
these restoration efforts will be done for the two regions.
Introduction ____________________
The invasion of exotic annual grasses into native vegetation has been proceeding in some areas of the world for the
last 100 years (D’Antonio and Vitousek 1992), and is
currently affecting two major regions of the United States.
The invasion of cheatgrass (Bromus tectorum) into big
sagebrush (Artemisia tridentata) shrub-grasslands of the
Intermountain Region is by now well documented (Mack
1981; McArthur and others 1990; Monsen and Kitchen
1994), although few large-scale solutions to restore native
species have been implemented (Pellant 1994). A parallel
but more recent invasion is occurring with the invasion of
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Edith B. Allen is Professor, Department of Botany and Plant Sciences,
University of California Riverside, Riverside, CA 92521-0124, U.S.A., e-mail:
eallan@citrus.ucr.edu
USDA Forest Service Proceedings RMRS-P-31. 2004
red brome (Bromus madritensis ssp. rubens) and other
species of brome grasses, wild oats (Avena), barley (Hordeum), and annual fescues (Vulpia) into California sagebrush (Artemisia californica) shrublands of the coastal
sage scrub vegetation of California. The latter invasion is
so recent, an event that has been occurring over the past 25
to 30 years, that there is still little awareness that it is even
occurring, or that it represents a permanent shift in vegetation type from perennial shrubland to annual grassland
(Allen and others 1998; Minnich and Dezzani 1998). Yet
this invasion has such similarities with the better-studied
cheatgrass invasion that the parallel cannot be missed.
Through a comparative examination we can learn lessons
from cheatgrass invasion that may be applied to California
sagebrush invaded by annual brome and other grasses,
both what to expect and what control and restoration
measures may be implemented. The objective of this review
is to examine the similarities and differences of the two
invasions by examining the ecology of the species, the
climates, soils and disturbance regimes of the two regions.
I will discuss potential reasons for the recent invasion and
vegetation-type conversion of California sagebrush, and
the potential for restoration of this vegetation as compared
to big sagebrush.
Land Area Affected and Site
Descriptions ___________________
California sagebrush is one of several dominant shrubs
of the coastal sage scrub (CSS) vegetation that extends
along the coast from the San Francisco Bay area southward
into Baja California. The vegetation has its greatest extent
and inland distribution in southern California in Riverside
County and adjacent counties, where it is called inland
sage scrub or Riversidean sage scrub (Westman 1981). The
more inclusive term CSS will be used to include those areas
codominated by such species as California buckwheat
(Eriogonum fasciculatum) and brittlebush (Encelia
farinosa) that are also undergoing conversion to exotic
annual grasslands.
The CSS of southern California is among the most intensively human-impacted vegetation types in the United States.
For this reason it has become a focus for mitigation and
restoration (Bowler 1990), driven by the legal requirements
9
Allen
of the Endangered Species Act (ESA). Currently, Habitat
Conservation Plans under the ESA are underway in a
number of counties of southern California that have large
components of potential CSS, some with up to 160 threatened, endangered, and sensitive species (http://www.
ecoregion.ucr.edu). CSS has been subject to urbanization,
agriculture, and invasion by exotic annuals to such a large
extent that one estimate gives only 10 percent of the original
vegetation in good condition (Westman 1981). A more recent
estimate puts CSS loss at 66 percent (O’Leary and others
1992), but in either case it is a significant loss because CSS
covers only about 1.2 million ha of land area. The difference
between the estimates may lie in the inability to estimate
accurately how much of the natural potential vegetation is
CSS, and it may lie in the condition of the remaining
vegetation, which is often very weedy. Large expanses of
CSS have been invaded by Mediterranean annual grasses
and forbs (Freudenberger and others 1987), and in many
areas the shrub understory is dominated by exotic rather
than native annuals (Minnich and Dezzani 1998; Stylinski
and Allen 1999). The conversion of shrubland to Mediterranean annual grassland has occurred primarily in the last 25
to 30 years, with 90 percent loss in shrub cover close to urban
areas in western Riverside County and permanent replacement by annual grasses (Minnich and Dezzani 1998). Exotic
annual grasses invaded California native grasslands over
the past 200 years (Heady 1977), but the vegetation typeconversion of CSS to exotic annual grassland is a recent
phenomenon.
The big sagebrush-grasslands of the Great Basin, the
Columbia Plateau, and the Colorado Plateau have been
undergoing invasion of cheatgrass and other species for
more than 100 years (Billings 1990; Mack 1981; Young and
others 1972). I will rely on literature citations and use less
detail in explaining the cheatgrass conversion, as it is better
documented than the California situation, and was the
subject of two earlier Shrub Symposium volumes (McArthur
and others 1990; Monsen and Kitchen 1994). Some 18
million acres of Intermountain rangeland have been invaded by cheatgrass plus medusahead (Pellant and Hall
1994). In fact, cheatgrass is still invading new areas where
it has been sparse or absent, as in shrub-grasslands of the
Colorado Plateau in Utah (Gelbard and Belnap 2003).
The big sagebrush type-conversion does not have the
extensive conservation implications for endangered species
as does the CSS. There are, however, several notable restoration projects for endangered animals such as the blackfooted ferret and the sage grouse (Wrobleski and Kauffmann
2003; Wirth and Pyke 2003) in the big sagebrush-grasslands, as well as some endangered plant species such as
Silene spaldingii in Montana (Lesica 1997) and Gaura
neomexicana subsp. coloradensis in Wyoming (Munk and
others 2002). However, huge land areas of big sagebrush are
involved relative to the California CSS, and although certain
species are of concern, the area does not have the hundreds
of sensitive species that California has. This fact has implications for policy on restoration, which is driven by mitigation coupled with development in California, while restoration is more often driven by land productivity needs in the
big sagebrush-grassland regions.
10
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
Climate and Phenology __________
Cheatgrass and the Mediterranean grasses occur in semiarid, winter-wet, and summer-dry climates, but cheatgrass
occurs in areas with winter frost, while the Mediterranean
grasses are frost intolerant. Their precipitation regimes are
broadly overlapping, although red brome also occurs in cool
and/or less dry desert areas. Cheatgrass typically germinates in the fall in areas where few or no native species are
adapted as winter annuals, as it has frost-tolerant seedlings.
Cheatgrass is not persistent in the eastern edge of big
sagebrush-grassland where summer rains are more frequent than in the Great Basin. Warm season grasses make
their appearance in regions with summer rain, and winter
annuals cannot compete. In addition, the high basins and
high plains of regions to the east of the Great Basin have cold
winters that are not conducive to the survival of winter
annuals. In the sagebrush-grassland (1,500 m) of eastern
Wyoming, cheatgrass is a spring annual in most years that
senesces by early summer (Allen and Knight 1984).
The Mediterranean annual grasses also germinate with
the onset of fall rains, and are often phenologically earlier
than California native annuals and perennials (Chiariello
1989). The earlier phenology makes the exotic grasses highly
competitive with native species. Plant growth occurs during
the mild and rainy winter growing season in California, as
summers are dry. The following spring the grasses produce
seeds and senesce at the same time or earlier than native
annuals. In the Mediterranean, rainfall typically continues
later into the spring and early summer, giving an advantage
to native perennial species that persist into the dry summer.
Annual grasslands are considered early successional in the
Mediterranean, and the extent of annual grasslands is less
than in California (Jackson and Roy 1989). The annual grass
phenology coupled with the shorter rainfall season is cited as
the reason for persistence of the annual grasslands in
California (Jackson and Roy 1989). Another argument is
made below that soils may also contribute to the dominance
of Mediterranean grasses in California.
Soil Nitrogen and Phosphorus ____
Soil nutrients may also contribute to the abundance of
exotic annual grasslands in the Western United States. The
Mediterranean has primarily calcareous soils originating
from marine limestone deposits (Blondel and Aronson 1999),
while much of southern California has granitic soils (Norris
and Webb 1976). Both regions have volcanic outcroppings,
more extensive in northern California. The calcareous soils
of the Mediterranean are phosphorus fixing, with low extractable levels of P (Rabinovitch-Vin 1983). Grasslands
dominate in volcanic outcroppings that have higher levels of
soil P, but not in calcareous soils. For instance, the region
around the Sea of Galilee in Northern Israel is volcanic with
an annual grassland, while nearby calcareous soils are
dominated by Mediterranean shrublands (Rabinovitch-Vin
1983). The soils of southern California tend to have moderate to high P levels. Measured amounts in sites in San Diego
and Riverside Counties were 20 to 40 mg per kg bicarbonate
extractable P, and 700 to 1,000 mg per kg total P (Cannon
USDA Forest Service Proceedings RMRS-P-31. 2004
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
and others 1995; Nelson and Allen 1993; Padgett and others
1999).
Not only soils of southern California, but also much of
the Great Basin are relatively high in extractable P. A
survey of soil P from 90 locations across the Western United
States where big sagebrush (A. tridentata ssp. tridentata)
occurs, showed that most of the sites had levels of P from
11 to 20 mg per kg, with a mean of 15.3 (S.E. = 0.9). The
exceptions were sites on the edges of its distribution, such
as North Dakota and Baja California, that had under 10 mg
per kg (fig. 1). This indicates that big sagebrush is a species
of moderate to high P soils, but may also explain why it is
so invasible by exotic species that have their greatest extent
in soils rich in P in their home environment. Many invasive
species have agricultural origins, either from cropland or
pasture, and may be well adapted to soils naturally rich in
nutrients.
Some Great Basin soils are also relatively high in N, and
cheatgrass is highly responsive to N fertilizer (for example,
Yoder and Caldwell 2002). The soils of southwest Wyoming
sagebrush-grassland had 6 percent organic matter and 0.5
percent total Kjeldahl N, a surprisingly high amount for a
semiarid soil (Waaland and Allen 1985). The richness of
these soils may be explained by their geologic history. Some
are relicts of more mesic vegetation from the last glaciation,
while others are in sedimentary basins that in some cases
were Pleistocene lake bottoms (Fiero 1986). Exotic plant
invasion was highly correlated to soils high in P and N in
southern Utah (Bashkin and others 2003).
Allen
Nitrogen Deposition as a Cause of
Invasion _______________________
The relatively high soil nutrient levels may explain in part
why annual brome grasses have invaded big sagebrushdominated lands of the Intermountain Region in the past,
but not why the CSS of California has been recently invaded.
The urbanizing areas of southern California are also subject
to anthropogenic nitrogen deposition, which has been occurring for the past 40 years primarily as NOx from automobile
emissions (fig. 2). The adoption of the catalytic converter in
the mid-1980s caused a general decline in emissions, but
NOx is still high enough to cause elevated soil N (Padgett
and others 1999). NOx is converted to the plant-available
–
forms of HNO3 per NO3 in the atmosphere. In agricultural
+
regions, NH3 per NH4 is the more abundant form of deposited N, but it forms only about 10 percent of the total
deposited N in urban southern California, depending on the
proximity to agricultural sources (Fenn and others 2003;
Padgett and others 1999). Some of the fixed N is deposited as
wetfall, but up to 90 percent may be deposited as particulate
or ionic forms to plant and soil surfaces in the dry Mediterranean summers. The dryfall N accumulates on surfaces
until the following rainy season when it is leached into the
soil and taken up by plants. Levels of N deposition are high
in southern California, with 30 kg per ha per year of N in
shrublands (Bytnerowicz and others 1987) and up to 50 kg
per ha per year in higher elevation pine forest (Fenn and
others 2003). Only the Netherlands has higher levels, up to
+
90 kg per ha per year, mainly in the form of NH3 per NH4 .
These high levels have been experimentally shown to cause
loss of shrub cover in heathlands, with a concomitant increase in perennial native grasses (Bobbink and Willems
1987).
Soils in southern California CSS have up to 87 kg per ha
per year extractable N measured near urban areas with the
–
greatest atmospheric NO3 concentrations (Padgett and
others 1999). These high levels of N may also be the cause of
the recent exotic grass invasion (Allen and others 1998,
3500
3000
NOx (T/day)
2500
2000
1500
1000
500
Figure 1—Concentrations of bicarbonate
extractable soil P in sites dominated by Artemisia
tridentata ssp. tridentata (data from E. Allen and
M. Allen).
USDA Forest Service Proceedings RMRS-P-31. 2004
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1961
1937
0
Figure 2—Emissions of NOx in the South Coast
Air Basin of southern California, 1937 to 2000
(Alexis and others 2000).
11
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
Allen
Nitrogen and the Fire Cycle _______
The fire frequency has been reduced from 60- to 100-year
to 5-year intervals in big sagebrush-grassland invaded by
cheatgrass (Billings 1990; Whisenant 1990). This is largely
because cheatgrass forms a fine, continuous fuel that carries
fire from shrub to shrub, or from one bunchgrass to the next.
Bunchgrasses, like shrubs, are not as efficient at carrying
fire long distances (Whisenant 1990). The increase in fine
12
Shrubs
120
+N
-N
Percent cover
100
80
*
60
40
20
0
1996
1997
1998
1999
2000
2001
Exotic grasses
160
*
140
*
+N
120
-N
100
g/m2
2000). Like cheatgrass in the Great Basin, the Mediterranean brome grasses were introduced to California more than
100 years ago, and some species arrived in the mid-1700s
with the first Spanish missions (Heady 1977). Species such
as wild oats were abundant in the Central Valley by the mid1800s where extensive farming and grazing was already
underway. But the CSS was also grazed until the early
1900s, and was burned by ranchers to increase annual grass
cover (Burcham 1957). However, it appears to have been
resilient to permanent vegetation type conversion until the
1960s (Minnich and Dezzani 1998). The Forest Service Vegetation Type Map survey was done throughout California
during the 1930s, including some 100 sites in CSS of western
Riverside County. Minnich resurveyed these areas in the
early 1990s and found that some had lost up to 90 percent of
their shrub cover, especially those near urban areas (Minnich
and Dezzani 1998). Local botanists have also noted that the
diverse native shrubland, with its colorful understory of
spring flowers, has converted to monotypic exotic annual
grassland in the past 25 to 30 years (for example, Oscar Clark,
retired herbarium curator, and Andrew Sanders, herbarium
curator, University of California-Riverside).
The simplest explanation for type conversion under N
deposition would be that the exotic grasses are responsive
to elevated soil N to a greater extent than the native
shrubs, as occurred in the Netherlands (Bobbink and
Willems 1987), but the situation is more complex in this
semiarid climate. Three invasive species, Bromus rubens,
A. fatua, and Brassica geniculata, were all highly responsive to N fertilization, but three native shrub species were
also responsive, including California sagebrush (Padgett
and Allen 1999). Plant response to N in the greenhouse
alone did not explain the observed shifts from shrubland to
annual grassland, so field studies were initiated. A longterm fertilization experiment was initiated in CSS 40 km
south of Riverside, CA, in an area with low N deposition.
The site was burned by a wildfire in fall of 1993, and was
fertilized yearly thereafter with 60 kg per ha per year of
NH4NO3 (Allen and others 1998). Annual grasses responded
to N fertilizer with an increase in biomass during the wet
years, but there were no long-term changes in shrub cover
(fig. 3). If there are direct effects of N on shrubs they may
be even longer term than this experiment. For instance,
species shifts in shortgrass prairie fertilized with 100 kg N
annually for 4 years did not materialize until about 5 years
after the cessation of fertilization (Milchunas and Lauenroth
1995), so further observations of these plots will continue.
However, elevated grass biomass has another impact, which
is to increase the fire cycle, and N fertilization may have
more immediate effects than species shifts in this vegetation type by altering the fire cycle, as discussed next.
*
*
80
*
60
*
40
20
0
1994
1995
1996
1997
1998
1999
2000
2001
Figure 3—Responses of exotic annual grasses
and native shrubs to N fertilization, which was
begun in spring 1994 following a November fire in
1993 (data from Allen and others, in press).
fuel frequency with an increase in the fire cycle has occurred
in cheatgrass-invaded lands over the past 100 years, but has
only been anecdotally recognized in the past 20 years in CSS.
Short fire intervals cause conversion of shrubland to grassland (Calloway and Davis 1993), but short fire intervals can
only occur when there is sufficient fuel to carry a fire in
subsequent years, for example, following annual grass invasion. The normal fire interval in CSS is about 30 years
(Westman and O’Leary 1986). Urban parks in western
Riverside County, such as Box Springs Mountain and Mt.
Rubidoux, that were dominated by CSS only 20 years ago are
now largely annual grasslands because of fires that burn at
2- to 5-year intervals (Cione and others 2002; Minnich 1988).
Based on observations from aerial photos, fire frequency
probably increased in urban Riverside County in the mid1960s when persistent patches of annual grasses appeared
(Minnich, personal communication). However, even sites
that have not burned in 30 years have lost shrub cover, and
grass-dominated sites that burned or were disturbed several
decades ago have not recovered to their former diversity of
native shrubs and understory herbs (Freudenberger and
others 1987; Stylinski and Allen 1999). We can learn from
cheatgrass-dominated big sagebrush, where initial small
incursions of grass promoted fire, which produced a positive
feedback loop of grass and fire until an annual grassland
stabilized under a frequent fire cycle. Nitrogen deposition will
exacerbate the positive feedback by increasing the fine fuel.
USDA Forest Service Proceedings RMRS-P-31. 2004
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
Restoration ____________________
Restoration is driven by different needs in southern California CSS and in Intermountain sagebrush-grasslands.
Fire hazards and large-scale loss of productive grazing lands
are primary drivers for restoration in the Intermountain
Region, with enhancement of endangered species a secondary or regional issue, depending upon local species occurrence. Conversely, the large number of endangered species
coupled with rapid urban development in southern California has created enormous species conservation issues that
are in part addressed through mitigation restoration. The
fire hazard of CSS located adjacent to urban areas has also
increased as fine grass fuels invade, and control of grasses to
reduce flammability is a goal in these areas. Rangeland
productivity is rarely considered a revegetation goal in CSS
any more, because most of the remaining rangelands have
been or are being converted to conservation reserves. Where
rangeland productivity and fire control of big sagebrush is
the goal, revegetation plantings tend to focus on exotic
species, while revegetation in CSS nowadays is always done
with native plants because of the sensitive species issues.
Whether the objective is restoration for rare species or to
reduce the fire hazard, the first step is to control annual
grasses. A few of the techniques to control the grasses, and
their relative success in California and the Intermountain
Region, are compared below. These are primarily fire, grazing, herbicides, and mechanical control.
Controlled spring burning of exotic grasses has been used
successfully to restore perennial purple needlegrass (Nassella
[= Stipa] pulchra) grassland at the Santa Rosa Plateau, a
former ranch that is managed by The Nature Conservancy
as part of the Riverside County Multispecies Habitat Conservation Plan. While results of fire management from this
site are not in the published literature, the managers have
demonstrated that late summer or fall burning maintains
the annual grass cover, most likely because the seed density
is not reduced by fire. By contrast, spring burns before the
seeds have shattered enable a complete burn of the potential
seedbank. The seedbank of these species is not persistent for
many years, with 1 percent or less remaining in the soil after
fall germination, so that seedbank control is readily achieved
with spring fire (Young and others 1981).
Fire has not been used as a tool to restore weedy CSS
because the dynamics of plant response to fire are different
in CSS than in grassland. Perennial bunchgrasses resprout
immediately after a spring fire, but reseeding shrubs (most
species in CSS are reseeders following fire) must compete
with residual weed seeds in the seedbank. Prescribed fire
has also seldom been used as a tool for restoration of weedy
big sagebrush-grassland because native seed banks are
often depleted, and cheatgrass will become more abundant
with the removal of the shrub cover (Rasmussen 1994). Fire
was successfully used to restore native forbs, either seeded
or from the seedbank (Wirth and Pyke 2003; Wrobleski and
Kauffmann 2003), but these studies were done in areas with
low cheatgrass cover. Fires in weedy big sagebrush must
typically be seeded because of the lack of a native seedbank
(Rasmussen 1994), and similar measures would need to be
undertaken in weedy CSS.
USDA Forest Service Proceedings RMRS-P-31. 2004
Allen
Mechanical or herbicide removal of grasses followed by
shrub planting is frequently used in CSS restoration mitigation. For instance, grasses (red brome, ripgut brome [Bromus
diandrus], wild oats) were experimentally thinned by hand
2
to densities varying between 0 and 500 grasses per m .
Growth of planted California sagebrush was reduced even at
2
the lowest density, 25 grasses per m compared to the control
(Eliason and Allen 1997). In another study, control of grasses
by mowing or grass-specific herbicide enabled establishment of seeded shrub species, including California sagebrush, but weedy control plots had no shrub establishment
at all (Cione and others 2002). Similarly, native plants
establish poorly in dense stands of cheatgrass. However,
several studies have shown that exotic perennials originating from Asia may readily be seeded into cheatgrass, sometimes even if it has received little treatment to control the
annual grasses (Davis and Harper 1990; McArthur and
others 1990; Monsen and Turnipseed 1990).
Because of the reluctance to use fire in CSS, studies are
underway to control annual grasses and increase shrub and
native understory herbs. The study reported here took
advantage of the native seedbank, and did not include
seeding. The study site is at Lopez Canyon in the Western
Riverside County Multispecies Reserve, 55 km south of
Riverside, CA, in an area with relatively low N deposition.
The area was grazed by cattle until the mid-1980s when it
was transferred to Riverside County as mitigation habitat
for development of CSS vegetation in other parts of the
County. The site was burned by a wildfire in 1993, but has
a relatively high proportion of grasses in the understory. The
grass invasion may be related to the grazing history of the
site. Two treatments, plus an untreated control, were applied in each of three replicate, 1-ha plots: (1) sheep grazing
in March per April 1999, 2000, and 2001, and (2) Fusilade
(grass-specific) herbicide application in February/March
1999 and 2000. Sheep grazing was done by fencing each of
the 1-ha plots, and allowing 200 sheep to remain in each plot
for 48 hours. Grazing was done for 3 consecutive years
because grasses had already gone to seed the first 2 years,
and sheep did not consume the grasses until the third year
when more forage was available. Plots were arranged in a
randomized block design with the three treatments plus a
control in each of the three blocks. Percent cover data of
vegetation were collected in 0.5- by 1.0-m plots, 20 plots per
ha, in late April/early May of each year. No additional
treatments were applied in 2002, but percent cover data
were collected.
The exotic grass cover (mainly red brome, with some
ripgut brome, wild oats, and annual fescue) decreased with
the herbicide treatment in 1999 compared to control plots,
and decreased more in 2000 (fig. 4). No additional herbicide
was applied in 2001, but exotic grass cover was still significantly lower than in control plots. By 2002 there were no
significant differences among the treatments, but this was
the driest year on record in this region since recordkeeping
began in the State, only about 10 cm in an area that receives
28 cm.
The grazing treatment was not effective in controlling
grasses until 2001 (fig. 4). The grazing was done in March,
before the data were collected in each year, and showed that
the sheep did not consume grasses until 2001. This was the
13
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
Allen
Native forb
Exotic grass
35
60
Control
a
b
30
a
Fusilade
50
40
Percent cover
Percent cover
25
Grazing
b
a
30
a
a
b
20
20
15
b
a
10
c
b
10
a
a
a
5
a
a
ab
0
b
a
a
1999
2000
2001
2002
1999
2000
2001
Year
Exotic forb
a
2002
Native shrub
30
a
25
a
25
a
a
a
20
a
a
a
a
b
a
10
Percent cover
20
Percent cover
a
Year
30
15
a
0
a
a
a
15
a
a
a
a
a
a
a
10
a
b
5
5
a
0
0
1999
2000
2001
2002
1999
Year
2000
2001
2002
Year
Figure 4—Effects of sheep grazing, grass specific herbicide, and herbicide plus dethatching
on percent cover of native forbs, exotic forbs, and shrubs in a coastal sage scrub site in
southern California over 4 years. The 2002 growing season was extremely dry, with virtually
no germination of native forbs.
only year when the grasses had not produced seed by the
time the sheep were placed in the plots. This was not due to
poor management, but rather to timing of rains. The rains
came late in 1999 and 2000, and by the time grasses had
produced forage for sheep to graze, they had also produced
seeds. Timing of precipitation was more conducive to annual
grass grazing in 2002.
The native forbs consisted of some 20 to 30 annual and
perennial species each year, none of which was very abundant
14
individually. Their combined cover increased significantly
in the 2000 and 2001 growing seasons in response to the
herbicide treatment, but decreased with grazing. In 2002
there was virtually no forb growth at all due to the drought.
The sheep consumed the forbs preferentially, but to determine whether the sheep had positive or negative long-term
effects on the native forbs the plots will need to be observed
for additional years. Whether the native forbs can recover
from sheep grazing will depend on the longevity of their
USDA Forest Service Proceedings RMRS-P-31. 2004
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
seedbanks. This is likely quite long, as the annuals are
largely fire-following species that rest in the soil between the
30-year fire intervals in CSS.
The exotic forbs consisted mainly of storksbill (Erodium
cicutarium and E. brachythecium) with less than 2 percent
wild mustard (Brassica geniculata), but there were no significant differences among the treatments in any year.
Similarly, the native shrubs did not respond in cover to the
treatments. We examined the sample quadrats for germinating shrub seedlings, but did not observe any in response
to any of the treatments.
Overall, the study showed that native forbs increase in
cover following grass-specific herbicide, but there was no
improvement of shrub cover. Whether this requires seeding
or longer term treatment is unknown. Some of the shrubs
have improved germination with smoke or fire (Keeley
1991), so a fire treatment may be needed, but controlled fire
is currently unlikely because managers fear further loss of
the shrub canopy, as described above.
Restoration at Lopez Canyon will be difficult even without the added grass productivity that comes from elevated
soil N. A more difficult case of CSS restoration with high
soil N was done in Riverside under high N deposition in a
pure stand of exotic grasses. Mowing and herbicide treatments were used to control the grasses so seeded shrubs
could establish. There was no native seed in the seedbank,
as shown by absence of native plant emergence in weeded
but unseeded plots (Cione and others 2002). The addition of
mulch is a technique that can help reduce soil N and
improve grass establishment. A restoration study on a
pipeline disturbance with elevated soil N showed that bark
mulch was more effective than straw mulch at immobilizing soil N, and bark also promoted growth of larger California sagebrush plants (Zink and Allen 1998). However, mulching will probably not be a large-scale method of restoration
because transporting mulch to large areas is not cost
effective. Where grass biomass is high because of ambient
or elevated soil N levels, grass control methods are likely
more cost effective over large areas than mulch for reestablishment of native vegetation. Cheatgrass-invaded
lands may in fact become reduced in N over time due to
increased N turnover and leaching (Evans and others
2001), although no change in N dynamics was observed in
another study (Svejcar and Sheley 2001).
Conclusions: Vegetation Stability
and the Limits to Restoration _____
The stability of restored vegetation on lands invaded by
exotic annual grasses is still in question. California sagebrush and big sagebrush-dominated lands are invasible, at
least in part, because the soils are moderate to high in P and
N, and subject to N deposition in urbanized California.
Annual brome grasses tend to be highly responsive to
nutrients, and thus preadapted to these high soil nutrient
levels. If the natural vegetation is ecologically unstable and
subject to grass invasion, restored vegetation will likely
also be unstable. In contrast, some of the most stable
revegetated lands once dominated by cheatgrass are those
planted to exotic perennials such as forage kochia or crested
USDA Forest Service Proceedings RMRS-P-31. 2004
Allen
wheatgrass (Agropyron cristatum) in the Intermountain
region. Recent hopeful studies show that one native species, squirreltail (Elymus elymoides), colonizes cheatgrass
naturally, and preferentially allows establishment of sagebrush but not cheatgrass (Booth and others 2003). This
may open the door for restoration of big sagebrush, but a
similar solution for restoration of CSS is not at hand.
Bowler (1990) has stated that the real test of the success of
CSS restoration will come after a restored site has burned,
and then re-established again. One incident occurred in
replanted CSS in Riverside, where a late spring fire burned
annual grasses, but adjacent stands of planted shrubs did
not burn (Cione and others 2002). The fire did not burn
shrubs because of differences in fuel moisture, as grasses
were senescent and dry while the deeper rooted shrubs
stayed green later into the summer. The goals of a reduced
fire cycle and improved forage for grazing can likely be met
readily by use of suitable exotic species in cheatgrassinvaded lands, but the more difficult goal of restoration of
native plant species and animal habitat will not only
require more initial effort, but also long-term maintenance
to control invasion of annual grasses.
References _____________________
Allen, E. B.; Knight, D. H. 1984. The effects of introduced annuals
on secondary succession in sagebrush grassland, Wyoming. The
Southwestern Naturalist. 29: 407–421.
Allen, E. B.; Padgett, P. E.; Bytnerowicz, A.; Minnich, R. 1998.
Nitrogen deposition effects on coastal sage vegetation of southern
California. In: Arbaugh, M., ed. Proceedings of the international
symposium on air pollution and climate change effects on forest
ecosystems; 1996 February 5–9; Riverside, CA. Gen. Tech. Rep.
PSW-GTR-166. Berkeley, CA: U.S. Department of Agriculture,
Forest Service, Pacific Southwest Research Station: 131–140.
http://www.rfl.psw.fs.fed.us/pubs/psw-gtr-164/fulltext/allen/allen/
allen.html#anchor1473574.
Allen, E. B.; Sirulnik, A. G.; Egerton-Warburton, L.; Kee, S. N.;
Bytnerowicz, A.; Padgett, P. E.; Temple, P. J.; Fenn, M. E.; Poth,
M. A.; Meixner, T. [In press]. Air pollution and vegetation change
in southern California shrublands. In: Kus, B. E.; Beyers, J. L.,
eds. Planning for biodiversity in southern California: bringing
research and management together: proceedings of the symposium; 2000 February 29–March 3; Pomona, CA. Riverside, CA:
U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station.
Bashkin, M.; Stohlgren, T. J.; Otsuki, Y.; Lee, M.; Evangelista, P.;
Belnap, J. 2003. Soil characteristics and plant exotic species
invasions in the Grand Staircase-Escalante National Monument,
Utah, USA. Applied Soil Ecology. 22: 67–77.
Bobbink, R.; Willems, J. H. 1987. Increasing dominance of Brachypodium pinnatum (L.) Beauv. in chalk grasslands: a threat to a
species-rich ecosystem. Biological Conservation. 40: 301–314.
Bowler, P. A. 1990. Coastal sage scrub restoration. I. The challenge
of mitigation. Restoration and Management Notes. 8: 78–82.
Burcham, L. T. 1957. California range land: an historico-ecological
study of the range resource of California. Sacramento, CA: Division of Forestry, Department of Natural Resources.
Bytnerowicz, A.; Miller, P. R.; Olszyk, D. M.; Dawson, P. J.; Fox, C. A.
1987. Gaseous and particulate air pollution in the San Gabriel
Mountains of southern California. Atmospheric Environment.
21: 1805–1814.
Billings, W. D. 1990. Bromus tectorum, a biotic cause of ecosystem
impoverishment in the Great Basin. In: Woodwell, G. M., ed. The
earth in transition: patterns and processes of biotic impoverishment. New York: Cambridge University Press: 301–322.
Blondel, J.; Aronson, J. 1999. Biology and wildlife of the Mediterranean region. Oxford University Press.
15
Allen
Booth, M. S.; Caldwell M. M.; Stark, J. M. 2003. Overlapping
resource use in three Great Basin species: implications for community invasibility and vegetation dynamics. Journal of Ecology.
91: 36–48.
Bowler, P. A. 1990. Coastal sage scrub restoration. I. The challenge
of mitigation. Restoration and Management Notes. 8: 78–82.
Callaway, R. M.; Davis, F. W. 1993. Vegetation dynamics, fire, and
the physical environment in coastal central California. Ecology.
74: 1567–1578.
Cannon, J. P.; Allen, E. B.; Allen, M. F.; Dudley, L. M.; Jurinak, J. J.
1995. The effects of oxalates produced by Salsola tragus on the
phosphorus nutrition of Stipa pulchra. Oecologia. 102: 265–272.
Chiariello, N. R. 1989. Phenology of California grasslands. In:
Huenneke, L. F.; Mooney, H. A., eds. Grassland structure and
function: California annual grassland. Dordrecht: Kluwer Academic Publishers: 47–58.
Cione, N. C.; Padgett, P. E.; Allen, E. B. 2002. Restoration of a native
shrubland impacted by exotic grasses, frequent fire and nitrogen
deposition in southern California. Restoration Ecology. 10: 376–384.
D’Antonio, C. M.; Vitousek, P. M. 1992. Biological invasions by
exotic grasses, the grass/fire cycle, and global change. Annual
Review of Ecology and Systematics. 3: 3–87.
Davis, J. N.; Harper, K. T. 1990. Weedy annuals and establishment
of seeded species on a chained juniper-pinion woodland in central
Utah. In: McArthur, E. D.; Romney, E. M.; Smith, S. D.; Tueller,
P. T., eds. Proceedings—symposium on cheatgrass invasion,
shrub die-off, and other aspects of shrub biology and management. Gen. Tech. Rep. INT-GTR-313. Ogden, UT: U.S. Department of Agriculture, Forest Service. Intermountain Research
Station: 71–79.
Eliason, S. A.; Allen, E. B.. 1997. Competition as a mechanism for
exotic grass persistence following conversion from native shrubland. Restoration Ecology. 5: 245–255.
Evans, R. D.; Rimer, R.; Sperry, L.; Belnap, J. 2001. Exotic plant
invasion alters nitrogen dynamics in an arid grassland. Ecological Applications. 11: 1301–1310.
Fenn, M. E.; Baron, J. S.; Allen, E. B.; Rueth, H. M.; Nydick, K. R.;
Geiser, L.; Bowman, W. D.; Sickman, J. O.; Meixner, T.; Johnson,
D. W. 2003. Ecological effects of nitrogen deposition in the
Western United States. Bioscience.
Fiero, B. 1986. Geology of the Great Basin. Reno: University of
Nevada Press.
Freudenberger, D. O.; Fish, B. E.; Keeley, J. E. 1987. Distribution
and stability of grasslands in the Los Angeles basin. Bulletin of
the Southern California Academy of Sciences. 86: 13–26.
Gelbard J. L.; Belnap J. 2003. Roads as conduits for exotic plant
invasions in a semiarid landscape. Conservation Biology. 17:
420–432.
Heady, H. F. 1977. Valley grasslands. In: Barbour, G.; Majors, J.,
eds. Terrestrial vegetation of California. New York: Wiley &
Sons: 491–514.
Huenneke, L. F.; Mooney, H. A., eds. 1989. Grassland structure and
function: California annual grassland. Dordrecht: Kluwer Academic Publishers.
Jackson, L. E.; Roy, J. 1989. Comparative ecology of annual grasses:
native versus Californian habitats. In: Huenneke, L. F.; Mooney,
H. A., eds. Grassland structure and function: California annual
grassland. Dordrecht: Kluwer Academic Publishers: 81–92.
Keeley, J. E. 1991. Seed germination and life history syndromes in
the California chaparral. The Botanical Review. 57: 81–116.
Lesica, P. 1997. Demography of the endangered plant, Silene
spaldingii (Caryophyllaceae) in northwest Montana. Madroño.
44: 347–358.
Mack, R. N. 1981. Invasion of Bromus tectorum into Western North
America: an ecological chronicle. Agro-Ecosystems. 7: 145–165.
McArthur, E. D.; Blauer, A. C.; Stevens, R. 1990. Forage kochia
competition with cheatgrass in Central Utah. In: McArthur, E. D.;
Romney, E. M.; Smith, S. D.; Tueller, P. T., eds. Proceedings—
symposium on cheatgrass invasion, shrub die-off, and other
aspects of shrub biology and management. Gen. Tech. Rep. INTGTR-313. Ogden, UT: U.S. Department of Agriculture, Forest
Service Intermountain Research Station: 56–66.
McArthur, E. D.; Romney, E. M.; Smith, S. D.; Tueller, P. T., eds.
1990. Proceedings—symposium on cheatgrass invasion, shrub
die-off, and other aspects of shrub biology and management. Gen.
16
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
Tech. Rep. INT-GTR-313. Ogden, UT: U.S. Department of Agriculture, Forest Service Intermountain Research Station.
Milchunas, D. G.; Lauenroth, W, K. 1995. Inertia in plant community structure—state changes after cessation of nutrient-enrichment stress. Ecological Applications. 5: 452–458.
Minnich, R. A. 1988. The biogeography of fire in the San Bernardino
Mountains of California. University of California Publications,
Geography. 28: 36–40.
Minnich, R. A.; Dezzani, R. J. 1998. Historical decline of coastal sage
scrub in the Riverside-Perris Plain, California. Western Birds.
29: 366–391.
Monsen, S. B.; Kitchen, S. G. 1994. Proceedings—ecology and
management of annual rangelands. Gen. Tech. Rep. INT-276.
Ogden, UT: U.S. Department of Agriculture, Forest Service,
Intermountain Research Station.
Monsen, S. B.; Turnipseed, D.; 1990. Seeding forage kochia onto
cheatgrass-infested rangelands. In. McArthur, E. D.; Romney, E. M.;
Smith, S. D.; Tueller, P. T., eds. Proceedings—symposium on
cheatgrass invasion, shrub die-off, and other aspects of shrub
biology and management. Gen. Tech. Rep. INT-GTR-313. Ogden,
UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station: 66–71.
Munk, L. M.; Hild, A. L.; Whitson, T. D. 2002. Rosette recruitment
of a rare endemic forb (Gaura neomexicana subsp coloradensis)
with canopy removal of associated species. Restoration Ecology.
10: 122–128.
Nelson, L. L.; Allen, E. B. 1993. Restoration of Stipa pulchra
grasslands: effects of mycorrhizae and competition from Avena
barbata. Restoration Ecology. 1: 40–50.
Norris, R. M.; Webb, R. W. 1976. Geology of California. New York:
John Wiley & Sons.
Padgett, P. E.; Allen, E. B. 1999. Differential responses to nitrogen
fertilization in native shrubs and exotic annuals common to
Mediterranean coastal sage scrub of California. Plant Ecology.
144: 93–101.
Padgett, P. E.; Allen, E. B.; Bytnerowicz, A.; Minnich, R. A. 1999.
Changes in soil inorganic nitrogen as related to atmospheric
nitrogenous pollutants in Southern California. Atmospheric Environment. 33: 769–781.
Pellant, M. 1994. History and applications of the Intermountain
greenstripping program. In: Monsen, S. B.; Kitchen, S. G. Proceedings—ecology and management of annual rangelands. Gen.
Tech. Rep. INT-276. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Research Station: 63–68.
Pellant, M.; Hall, C. 1994. Distribution of two exotic grasses on
Intermountain rangelands: status in 1992. In: Monsen, S. B.;
Kitchen, S. G. Proceedings—ecology and management of annual
rangelands. Gen. Tech. Rep. INT-276. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research
Station: 109–112.
Rabinovitch-Vin, A. 1983. Influence of nutrients on the composition
and distribution of plant communities in Mediterranean-type
ecosystems of Israel. In: Kruger, F. J.; Mitchell, D. T.; Jarvis, J.
U. M., eds. Mediterranean-type ecosystems: the role of nutrients.
Berlin: Springer-Verlag: 74–85.
Rasmussen, 1994. Prescribed burning considerations in sagebrush
annual grass communities. In: Monsen, S. B.; Kitchen, S. G. 1994.
Proceedings—ecology and management of annual rangelands.
Gen. Tech. Rep. INT-276. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station: 69–74.
Stylinski, C. D.; Allen, E. B. 1999. Lack of native species recovery
following severe exotic disturbance in southern California
shrublands. Journal of Applied Ecology. 36: 544–554.
Svejcar, T.; Sheley, R. 2001. Nitrogen dynamics in perennial- and
annual-dominated arid rangeland. Journal of Arid Environments. 47: 33–46.
Young, J. A.; Evans, R. A.; Raguse, C. A.; Larson, J. R. 1981. Germinable seeds and periodicity of germination in annual grasslands.
Hilgarida. 49: 1–37.
Waaland, M. E.; Allen, E. B. 1987. Relationships between VA
mycorrhizal fungi and plant cover following surface mining in
Wyoming. Journal of Range Management. 40: 271–276.
Westman, W. E.; O’Leary, J. F. 1986. Measures of resilience: the
response of coastal sage scrub to fire. Vegetatio. 65: 179–189.
USDA Forest Service Proceedings RMRS-P-31. 2004
Restoration of Artemisia Shrublands Invaded by Exotic Annual Bromus...
Whisenant, S. 1990. Changing fire frequencies on Idaho’s Snake
River Plains: ecological and management implications. In:
McArthur, E. D.; Romney, E. M.; Smith, S. D.; Tueller, P. T., eds.
1990. Proceedings—symposium on cheatgrass invasion, shrub
die-off, and other aspects of shrub biology and management. Gen.
Tech. Rep. INT-GTR-313. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station: 4–10.
Wirth, T. A.; Pyke, D. A. 2003. Restoring forbs for sage grouse
habitat: fire, microsites, and establishment methods. Restoration Ecology. 11: 370-377.
Wrobleski, D. W.; Kauffmann, J. B. 2003. Initial effects of prescribed
fire on morphology, abundance, and phenology of forbs in big
USDA Forest Service Proceedings RMRS-P-31. 2004
Allen
sagebrush communities in southeastern Oregon. Restoration
Ecology. 11: 82–90.
Yoder, C.; Caldwell, M. M. 2002. Effects of perennial neighbors and
nitrogen pulses on growth and nitrogen uptake by Bromus
tectorum. Plant Ecology. 158: 77–84.
Young, J. A.; Evans, R. A.; Rageuse, C. A.; Larson, J. R. 1981.
Germinable seeds and periodicity of germination in annual grasslands. Hilgardia. 49: 1–37.
Zink, T. A.; Allen, M. F. 1998. The effects of organic amendment on
the restoration of a disturbed coastal sage scrub habitat. Restoration Ecology. 6: 52–58.
17
Comparing Phenotype and
Fitness of Native, Naturalized,
and Invasive Populations of
Downy Brome (Cheatgrass,
Bromus tectorum)
C. Lynn Kinter
Richard N. Mack
Abstract: The Eurasian grass downy brome (cheatgrass, Bromus
tectorum L.) was introduced into arid and semiarid bunchgrass and
shrub communities in New Zealand and North America over 100
years ago, but has strikingly different histories in these ranges. On
New Zealand’s South Island, it persists at low levels, while in
Western North America, it dominates vast areas. Inherent high
fitness in founder populations may have contributed to the invasive
character of North American populations, while low fitness may
have precluded a New Zealand invasion. In four common greenhouse environments, 62 populations from Western North America,
New Zealand, and the native range in Western Europe were
compared for 15 phenotypic and fitness traits (including height,
days to flowering, vegetative biomass). Significant differences among
source locations were evident for most traits. North American
populations were typically most vigorous, followed by European,
and lastly New Zealand populations. North American plants flowered earlier, and New Zealand plants later, than those from the
native range. Differences in levels of fitness between the sources of
founder populations have likely contributed to radically different
histories of downy brome in two of its new ranges. Our research has
important implications for screening and predicting invaders and
controlling invasive species.
Downy brome, or cheatgrass (Bromus tectorum), is a
cleistogamous, annual C3 grass native to Eurasia and northern Africa. It was introduced from Europe (Novak and Mack
2001) into both North America and New Zealand over 100
years ago (Kirk 1869; Mack 1981), but has radically different histories in the new ranges. In Western North America,
it dominates vast areas formerly dominated by shrubs and
bunchgrasses (Mack 1981). On these rangelands, it changes
the structure of the community, often comprising over 90
percent of the vegetative cover. It decreases forage values,
causes loss of native biodiversity, and increases fire frequency, erosion, and siltation. In contrast, in the central
part of New Zealand’s South Island, downy brome is naturalized at low levels in shrub and bunchgrass communities
(Connor 1964; Williams 1980). There, it persists primarily
on disturbed sites, such as roadsides or trampled areas, but
does not become dominant over large acreages.
These performance differences might typically be attributed to environmental differences between the two introduced ranges. However, in the region of New Zealand where
downy brome has persisted for decades, the physiognomies
of the native plant communities, climate, and recent fire and
grazing history are similar to those features in Western
North America where downy brome has invaded heavily
(Mack 1986; Wardle 1991). Even two species that commonly
attack downy brome in its native range—head smut fungus
(Ustilago bullata) and bird cherry-oak aphid (Rhopalosiphum
padi)—have been known for at least a century in both
introduced ranges (Lowe 1961; Pergande 1904). Consequently, we hypothesized that differences in performance of
downy brome in these new ranges may be explained by
differences in the founder genotypes, rather than differences in the environments in the new ranges.
We assessed the potential for different performance among
downy brome from three ranges: the portion of its native
range in Western Europe that served as the donor region for
both North America and New Zealand (Novak and Mack
2001), and the new ranges in Western North America and
the South Island of New Zealand. By growing all populations
in common greenhouse environments, we investigated genetically based differences in phenotype, vigor, and fitness
among these three ranges.
Methods _______________________
Sample Collection
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
C. Lynn Kinter is a Botanist and Ph.D. Candidate, and Richard N. Mack
is a Professor of Botany, School of Biological Sciences, Washington State
University, Pullman, WA 99164-4236. FAX: (509) 335-3184; phone: (208) 3734381; e-mail: lkinter@fs.fed.us
18
In the native range, we based our collecting efforts on
Novak and Mack’s (2001) allozyme research. They found
that the widespread genotypes in Western North America
were most closely related to those in Europe, and less closely
related to genotypes in Eastern North America, Asia, and
Northern Africa. They deduced that the cheatgrass genotypes, which were the early founders that spread widely
USDA Forest Service Proceedings RMRS-P-31. 2004
Comparing Phenotype and Fitness of Native, Naturalized, and ...
across Western North America, were from Central and
Western Europe, probably around Slovakia, Czech Republic, or the eastern part of Germany. For this study, we
sampled heavily in that region, as well as north into Sweden
and west across The Netherlands, where Novak and Mack
did not sample but where many immigrants and shipments
left for the New World. We also included samples from Spain
and Italy. In the native range, downy brome is found at low
levels in native plant communities, such as stabilized dunes,
and on disturbed sites, such as railroad beds, farmsteads,
and cereal grain fields.
In Western North America, we sampled three populations
each in the vicinity of six founder locations identified by
Novak and Mack (2001)—from Nevada and Utah north to
British Columbia. In New Zealand, founder information is
not known, so we sampled three populations in the region
around Lake Pukaki and Bendigo (Canterbury and Otago
Counties) where herbarium specimens have been collected
since the 1930s. In total, we sampled 62 populations: 41 in
Europe, 18 in Western North America, and three in New
Zealand. From each population, we randomly collected 30
individuals, and ultimately chose six of those at random for
use in experiments.
Common Greenhouse Experiments
In December 2000, we planted seeds from six individuals
from each of 62 populations in four common greenhouse
environments: control, low water, low nutrients, and low
temperature. These seeds were one generation grown out
from the wild to minimize maternal effects (Roach and
Wulff 1987; Schaal 1984). No mortality occurred during the
grow-out phase, and all parent plants produced numerous
seeds for experiments. In the experimental phase, seeds
were sown in standard potting medium (60 percent peat, 20
percent pumice, 20 percent sand; pH 6.9–7.0) in individual
15-cm fiber pots on greenhouse benches under ambient
light. Temperatures were 24/16 ∞C (12h/12h) for 6 weeks
after sowing, followed by a 9-week cold period at 4/2 ∞C
(10h/14h) to ensure vernalization. Plants in the control
environment were watered daily to container capacity, and
fertilized weekly with Peters 20:20:20 delivered in line
during watering. Each of the remaining experimental environments differed from the control in only one factor.
Plants in the low-water environment were watered to
container capacity when the soil moisture content of 10
randomly chosen pots had fallen 50 percent. Plants in the
low-nutrients environment received only the nutrients
initially in the potting medium and trace minerals in tap
water. Three weeks after the seeds were sown, plants in the
low-temperature environment were placed in an adjacent
greenhouse room for a cold period of 4/2 ∞C (10h/14h) that
was 3 weeks longer than in the other environments. With
the exception of plants in the low-temperature environment during these 3 weeks, plants were arranged in a
completely randomized design, and rotated and re-randomized weekly to minimize block and bench effects. Water
and fertilizer treatments continued until all plants had
senesced. Each plant’s aboveground biomass was harvested when chlorophyll pigmentation in the panicles was
no longer visible—167 to 259 days after emergence—then
USDA Forest Service Proceedings RMRS-P-31. 2004
Kinter and Mack
dried at room temperature for 16 weeks. Throughout the
study, any extraneous plant material or collection bags
were autoclaved before disposal to prevent escape of genetic material.
While the plants were actively growing, we recorded days
to emergence, height of the tallest leaf every 30 days following emergence, width of the tallest leaf at 60 and 120 days
after emergence, tiller number at 60 days after emergence,
and days from emergence to flowering. At senescence, we
recorded days from flowering to senescence, number of
panicles, and first internode length of the tallest panicle.
After harvest and drying, first internode diameter of the
tallest panicle, aboveground vegetative biomass, panicle
biomass, and presence of glume hairs were recorded. Three
traits were also recorded before seeds were sown: weight of
50 seeds, seed length, and awn length.
Statistical Analyses
To assess trait differences among ranges, we compared
responses using Multivariate Analysis of Variance
(MANOVA) and Analysis of Variance (ANOVA) in Proc GLM
(SAS 2000) with a two-stage nested design: population
nested within range, and six replicates nested within population. Both population and range were fixed effects. Data
from each environment were analyzed separately due to
small but significant range by environment interactions. We
conducted a multivariate analysis of seven of the response
variables (height, leaf width, and tiller number at 60 days;
culm diameter, vegetative biomass, panicle biomass, days
from emergence to flowering) using Canonical Discriminant
Function Analysis (SAS 2000). All European and New
Zealand populations were included, but only the geographically central population from each of the six North American
founder sites was included to facilitate interpretation of the
scatterplot by reducing the amount of overlap in symbols. In
all comparisons, differences were determined to be significant when P < 0.05.
Results and Discussion __________
We found significant differences among ranges for nearly
every trait assessed in each of the four environments. For
example, in all but the early cold treatments, plants from
North America were significantly taller, those from New
Zealand were shorter, and those from Europe were intermediate (fig. 1). Similarly, North American plants had wider
leaves, New Zealand plants had narrower leaves, and European plants were intermediate (fig. 2). For aboveground
vegetative biomass at senescence, North American plants
had higher values and New Zealand plants had lower values
than did the European plants in each treatment; however,
differences were not significant in every comparison (fig.
3). The small number of populations from New Zealand
(n = 3) reduced test power and significant differences in
some cases.
For most of the remaining phenotypic traits we found
significant differences among ranges in each experimental
environment. In each of the four environments, North American plants flowered earliest and New Zealand plants latest
(fig. 4)—147 days for North America, 151 for Europe, and
19
Kinter and Mack
50
Comparing Phenotype and Fitness of Native, Naturalized, and ...
Control
9
North America
Europe
New Zealand
40
Control
North America
Europe
New Zealand
8
7
30
6
20
5
4
10
50
Low temperature
9
Low temperature
8
40
7
30
Leaf width (mm)
6
Height (cm)
20
10
50
Low nutrients
5
4
9
Low nutrients
8
40
7
30
6
20
5
4
10
50
Low water
9
Low water
8
40
7
30
6
20
5
4
10
30
60
90
120
Days after emergence
Figure 1—Mean height (cm) of downy brome
populations from North America, Europe, and
New Zealand at 30-day intervals after
emergence in each of four test environments.
Error bars indicate standard error.
20
150
60
120
Days after emergence
Figure 2—Mean leaf width (mm) of downy
brome populations from North America,
Europe, and New Zealand at 60-day intervals
after emergence in each of four test environments. Error bars indicate standard error.
USDA Forest Service Proceedings RMRS-P-31. 2004
Comparing Phenotype and Fitness of Native, Naturalized, and ...
Kinter and Mack
Aboveground vegetative biomass (g)
14
12
10
8
a
ab
6
a
b
b
b
a
4
a
b
ab
b
2
c
0
Control
Low
Low nutrients
Low water
Figure 3—Aboveground vegetative biomass for
downy brome populations from North America
(black), Europe (gray), and New Zealand (white) in
each of four environments. Error bars indicate
standard error. Within environments, differing letters
indicate significant differences (P < 0.05) among
ranges.
Emergence to flowering time (days)
160
140
120
100
a
b
b
a
b
c
a
b
c
a
b
c
80
60
40
Control
Low
Low nutrients
Low water
Figure 4—Time from emergence to flowering for
downy brome populations from North America
(black), Europe (gray), and New Zealand (white) in
each of four environments. Error bars indicate
standard error. Within environments, differing letters
indicate significant differences (P < 0.05) among
ranges.
156 days for New Zealand—when averaged across the four
environments. Not only are these differences statistically
significant, they are likely to be ecologically significant,
because previous work (Mack and Pyke 1983, 1984; Rice and
others 1992) has shown that a few days difference in flowering time can allow seed filling in arid environments.
Though population-level variability is not the key focus of
this study, we note that for nearly every trait assessed, we
found significant differences among populations within a
single range. This parallels findings by other researchers
working on Western North American populations and measuring traits such as biomass and seed dormancy (Meyer
and Allen 1999; Rice and Mack 1991a,b). Additionally, we
USDA Forest Service Proceedings RMRS-P-31. 2004
found less phenotypic variability among North American
and New Zealand populations than European populations,
as would be expected with the restricted genetic variability
following founder events based on a small number of introduced genotypes (Barrett and Husband 1990).
In analyzing all phenotypic traits simultaneously using
Canonical Discriminant Function Analysis in SAS, we found
that the centroids for each range were significantly different
from each other at P < 0.0001 (fig. 5). In each of the four
environments, the introduced phenotypes are essentially a
subset of the native phenotypes. North American phenotypes most closely match those from Austria, Slovakia, and
Southeastern Germany, while New Zealand phenotypes
mostly closely match those from Northwestern Germany
and The Netherlands—areas that lie geographically to the
northwest of the North American matches.
While plants from the three ranges clearly differ in phenotype and vigor, fitness differences may occur as well. We
found differences in mortality during a single time period
when plants in the low water treatment became extra dry in
a sudden spell of very hot days. North American plants had
a significantly lower death rate, 8.33 ± 2.66 percent, compared to European and New Zealand plants at 22.95 ± 2.69
percent and 27.78 ± 10.60 percent, respectively. A set of
experimental stress tests under different environments may
show that North American plants typically have higher
survival. They may also be more water use efficient, and we
plan to compare water use efficiency among the ranges using
carbon isotope analysis.
To further assess fitness, we are conducting seed germination trials and cleaning seeds from panicles to get a measure
of seed biomass produced in the four treatments. We will use
these two measures to calculate relative fitness values for
each population. Although these data are still being collected, some insight may be found in a previous analysis we
conducted in a single environment with one North American
population from Smoot Hill, WA, and 10 populations from
Central and Western Europe. In this pilot study, the North
American plants ranked second for relative fitness between
populations from first-ranked Bratislava, Slovakia, and
third-ranked Vienna, Austria. Plants from these two European populations were among those that displayed phenotypes most similar to those from North America in the four
greenhouse environments, and which were most genetically
similar based on allozyme data (Novak and Mack 2001). If
the pattern found in the pilot study corresponds to the larger
test in four environments, we expect plants from Western
North America to rank among those with highest fitness
from the native range.
Conclusions ____________________
To summarize our findings, plants from the invasive
range in North America were typically largest and most
vigorous, followed by those from the native European range,
and lastly those from the naturalized New Zealand range.
These and other phenotypic differences we have documented
among ranges indicate that genetic differences, rather than
environment alone, explain many performance differences
between native and introduced populations. Our findings
highlight the importance of studying representative genotypes
21
Kinter and Mack
Comparing Phenotype and Fitness of Native, Naturalized, and ...
Can
2
Control
e
e
Z
Z
e
2
Z
Z
A
A
e e
e
Z Z
e
e
e eAA
e
A
A
e e
e
e
e
Z e Ze eee e
A
e
e A
Ze
e Z e
eA
e
A
A
e Z
e e e e
e e A
A
e
e eZ
e e
e
e
A
ee e Z
e
Z
A
A
e e
eA
e
A
A ee e
e e
e
e
Z ee e e
eee
e
e eeeeeeAe e A
e
ee
eZ e
ee
e
ee
e
e
A
e ee
ee e e
e
e
eee
e eee
e
eA
ee
e
e
e
e A
e
e
e
ee
e
e
e
e
ee e
ee
e
A
e
e
eee e ee
ee
A
ee e ee
e
eee
e
e
e
e
e
e
ee
e e e e
e
e
e
e
ee
e
e
e
e e e
e e
e
e
e
e
e
A
e
A
e
e
e
e
e
e
e
e
Z
1
A
0
e
e
-1
-2
e
e
-3
A
-3
-2
-1
0
1
2
3
Can
1
Figure 5—Canonical Discriminant Function Analyses of seven phenotypic measurements
of downy brome in the control environment. Centroids of each range are indicated by large
letters: New Zealand (Z), Europe (e), and North America (A).
from across a species’ range in screening for invasive potential, testing for effectiveness of herbicides and biocontrol
agents, or similar studies. These results have ramifications
for quarantine regulations because, at present, an introduced species is often not considered to be a threat if it has
not been a problem in the past. Our study illustrates that the
outcome of an introduction can be very different based on
random genetic sampling in the native range. Serious ecological consequences may have occurred in New Zealand if
favorable habitats on the South Island had gained plants
from the populations that we unfortunately gained in Western North America.
22
Acknowledgments ______________
We thank J. R. Alldredge, M. A. Evans, and M. S. Minton for
statistical assistance; R. S. Bussa, C. A. Cody, M. A. Darbous,
M. Erneberg, S. K. Foster, G. K. Radamaker, B. D. Veley, and
K. D. Walker for technical assistance; and J. Baar, L. Barkan,
C. E. Hellquist, A. Klaudisova, N. L. Mack, H. C. Prentice,
R. R. Pattison, K. D. Rode, T. Sebastia-Alvarez, A. Tudela,
and L. J. van der Ent for seed collection assistance. We acknowledge helpful discussions with R. A. Black, L. D. Hufford,
M. Morgan, S. J. Novak, P. S. Soltis, and M. S. Webster. Funding sources include Betty W. Higinbotham and Noe
Higinbotham Research Awards, Hardman Foundation Award,
USDA Forest Service Proceedings RMRS-P-31. 2004
Comparing Phenotype and Fitness of Native, Naturalized, and ...
Sigma Xi—Scientific Research Society Grant-in-Aid-of-Research, and Washington State University College of Sciences
and School of Biological Sciences.
References _____________________
Barrett, S. C. H.; Husband, B. C. 1990. The genetics of plant
migration and colonization. In: Brown, A. H. D.; Clegg, M. T.;
Kahler, A. L.; Weir, B. S., eds. Plant population genetics, breeding, and genetic resources. Sunderland: Sinauer Associates.
Connor, H. E. 1964. Tussock grassland communities in the Mackenzie
Country, South Canterbury, New Zealand. New Zealand Journal
of Botany. 2: 325–351.
Kirk, T. 1869. On the naturalized plants of New Zealand, especially
with regard to those occurring in the Province of Auckland.
Transactions Proceedings of New Zealand Institute. 2: 131–146.
Lowe, A. D. 1961. The cereal aphid in wheat. New Zealand Wheat
Review. 8: 21–26.
Mack, R. N. 1981. Invasion of Bromus tectorum L. into Western
North America: an ecological chronicle. Agro-Ecosystems. 7.
Mack, R. N. 1986. Alien plant invasion in the Intermountain West:
a case history. In: Mooney, H. A.; Drake, J. A., eds. Ecology of
biological invasions of North America and Hawaii. New York:
Springer-Verlag: 192–213.
Mack, R. N.; Pyke, D. A. 1983. The demography of Bromus tectorum
L.: variation in time and space. Journal of Ecology. 71: 69–93.
Mack, R. N.; Pyke, D. A. 1984. The demography of Bromus tectorum:
the role of microclimate, predation and disease. Journal of Ecology. 72: 731–748.
USDA Forest Service Proceedings RMRS-P-31. 2004
Kinter and Mack
Meyer, S. E.; Allen, P. S. 1999. Ecological genetics of seed germination regulation in Bromus tectorum L. I. Phenotypic variance
among and within populations. Oecologia. 120: 27–34.
Novak, S. J.; Mack, R. N. 2001. Tracing plant introduction and
spread: genetic evidence from Bromus tectorum (cheatgrass).
Bioscience. 51: 114–122.
Pergande, T. 1904. On some of the aphides affecting grains and
grasses of the United States. USDA Division of Entomology
Bulletin. 44: 5–23.
Rice, K. J.; Black, R. A.; Radamaker, G.; Evans, R. D. 1992.
Photosynthesis, growth, and biomass allocation in habitat ecotypes
of cheatgrass (Bromus tectorum). Functional Ecology. 6: 32–40.
Rice, K. J.; Mack, R. N. 1991a. Ecological genetics of Bromus
tectorum. 1. A hierarchical analysis of phenotypic variation.
Oecologia. 88: 77–83.
Rice, K. J.; Mack, R. N. 1991b. Ecological genetics of Bromus
tectorum. 3. The demography of reciprocally sown populations.
Oecologia. 88: 91–101.
Roach, D. A.; Wulff, R. D. 1987. Maternal effects in plants. Annual
Review of Ecology and Systematics. 18: 209–235.
SAS. 2000. Version 8.1. Cary, NC: SAS Institute, Inc.
Schaal, B. A. 1984. Life-history variation, natural selection, and
maternal effects in plant populations. In: Dirzo, R.; Sarukhan, J.,
eds. Perspectives on plant population ecology. Sunderland: Sinauer
Associates: 188–207.
Wardle, P. 1991. Vegetation of New Zealand. Cambridge University
Press.
Williams, P. A. 1980. Vittadinia triloba and Rumex acetosella
communities in the semi-arid regions of the South Island. New
Zealand Journal of Ecology. 3: 13–22.
23
Root Growth of Apache Plume
and Serviceberry on
Molybdenum Mine Overburden
in Northern New Mexico
Mary I. Williams
Mohammed Tahboub
John T. Harrington
David R. Dreesen
April Ulery
Abstract: This study evaluated root growth of Apache plume
(Fallugia paradoxa) and Saskatoon serviceberry (Amelanchier
alnifolia) on overburden at the Molycorp mine near Questa, NM.
Container grown, 1-year-old seedlings were transplanted, and either fertilized or not fertilized at the time of planting in August
1995. Survival and shoot growth were monitored in 1996 and 2000.
Root density was determined at three distances (planes) from the
base of the plant at six depths in 2000. Fertilization effects on
serviceberry root densities were depth dependent. Total root density of unfertilized plants was greater than fertilized plants. Serviceberry total root density difºfered among planes (distances from
base of plant) between fertilization treatments. Fertilized plants
had fewer roots in the 10-cm plane relative to the 20- and 30-cm
planes. Fertilized plants of Apache plume had higher total root
densities at depths >20 cm within the 20-cm plane than plants not
fertilized at time of planting. Although overall performance in terms
of shoot growth was positive for both species, survival was generally
low and root density varied when fertilized at time of planting.
Factors including fertilizer characteristics, planting date, and site
conditions may have influenced species performance.
Introduction ____________________
New Mexico State and Federal laws require that mined
lands be reclaimed to support a designated postmining land
use (State of New Mexico 1999). Forestry as a postmining
land use has been encouraged by regulatory agencies since
the mid to late 1990s (Boyce 1999). Metal mines within the
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Mary I. Williams and John T. Harrington are Senior Research Assistant
and Associate Professor and Superintendent, Mora Research Center, New
Mexico State University, P.O. Box 359, Mora, NM 87732, U.S.A., e-mail:
maiwill@nmsu.edu. Mohammed Tahboub and April Ulery are Graduate
Research Assistant and Assistant Professor, Department of Agronomy and
Horticulture, New Mexico State University, Las Cruces, NM, U.S.A. David R.
Dreesen is an Agronomist, U.S. Department of Agriculture, Natural Resources and Conservation Service, Plant Materials Center, Los Lunas, NM
87031, U.S.A. FAX: (505) 387-9012.
24
mountainous Western United States including the Molycorp
molybdenum mine near Questa, NM, are selecting forestry
as a postmine land use to re-establish a plant community
comparable to adjacent native vegetation.
Open-pit mining at the molybdenum mine generated over
300 million metric tons of overburden from 1965 until 1983.
The overburden exists in piles that range in altitude from
2,400 to 3,000 m and are composed of mixed igneous rocks
(rhyolite and andesite) and black andesite and aplite materials (referred to as neutral rock). Mine overburden can be
low in organic matter, soil microorganisms, and plant nutrients such as nitrogen and phosphorus and can lack soil
structure and texture that are important to soil fertility and
water holding capacity (Allen 1989; Feagley 1985). These
properties make it challenging for plants to establish, thus
overburden or plant amendments are often recommended
(Brown and others 1996; Gardiner 1993).
Research studies on the survival of directly transplanted
trees and shrubs on the overburden began in the early 1990s
(Harrington and others 2001a,b,c). These studies indicated
that the overburden was suitable to support plant life, in
terms of transplant survival. In most instances, however,
shoot growth was limited. Other research has shown that
shoot growth of newly transplanted tree and shrub seedlings
can benefit from supplemental fertilization in both minimally disturbed (Fan and others 2002; Houle and Babeux
1994; Walker 1999a) and drastically disturbed ecosystems
(Fisher and others 1983; Voeller and others 1998; Walker
1999b). Growth response to fertilizer application, however,
can be both species specific (Voeller and others 1998; Houle
and Babeux 1994) and site specific (Gleason and others
1990).
The ability of transplanted seedlings to establish new
roots into a planting medium is essential to their survival
and subsequent growth. Plants with a well-established root
system add to the success of postmine revegetation, control
erosion, and improve overall postmine conditions. Most
fertilization studies, however, focus primarily on shoot growth
characteristics and disregard belowground growth. Early
plantings of ponderosa pine (Pinus ponderosa) at the Molycorp
mine were evaluated for root growth after 6 years of growth.
Rooting depth of some plants extended beyond 2 m in the
USDA Forest Service Proceedings RMRS-P-31. 2004
Root Growth of Apache Plume and Serviceberry on Molybdenum Mine Overburden...
overburden (Harrington, unpublished data). This observation led to questions regarding growth allocation of directly
transplanted trees and shrubs at the Molycorp mine with
fertilization treatments.
This study evaluated the influence of fertilization at time
of planting on root distribution and growth of two native
woody shrubs, Saskatoon serviceberry (Amelanchier
alnifolia) (hereafter referred to as serviceberry) and Apache
plume (Fallugia paradoxa), transplanted into overburden at
the Molycorp Questa Mine. Root densities and distributions
were described at six depths and three distances from the
base of each plant to examine the effects of fertilizer applications at the time of planting.
Materials and Methods ___________
Site Description
Molycorp, Inc., molybdenum mine is located near Questa,
NM, in the Red River Canyon and produces the largest
amount of molybdenum in the State (Schilling 1965; State of
New Mexico 1999). Overburden from open-pit mining was
deposited on the steep mountainsides of Red River Canyon.
The overburden piles are highly heterogeneous, consisting
of parent materials with a range of acidic and soluble salt
levels (Steffan and Kirsten 1995).
Vegetative communities surrounding the mine consist of
coniferous forests dominated by ponderosa pine, mixed conifer (Douglas fir [Pseudotsuga menziesii] and limber pine [P.
flexilis]), and spruce-fir (Engelmann spruce [Picea engelmannii] and white fir [Abies concolor]) stands (Harrington
and Wagner 1994). Naturally formed alteration scars occurring in acidic (pH = 1.8 to 3.5) materials, occur at the mine
and throughout the Red River Canyon (Meyer and
Leonardson 1990; Steffan and Kirsten 1995). Plants from
adjacent communities have established in the periphery of
these scars (Wagner and Harrington 1994).
Two planting sites (Blind Gulch and Spring Gulch) were
located on terraced portions of overburden piles at the
molybdenum mine. Spring Gulch (2,780 m) is composed
primarily of neutral rock with an average pH of 7.7, electrical conductivity (EC) of 0.5 dS/m, and coarse fragment
fraction (content) of 69 percent. Blind Gulch (2,860 m)
consists of both acidic and neutral overburden materials.
Chemical composition of the overburden across the planting
area at Blind Gulch was varied. Average values for pH, EC,
and coarse fragment fraction (content) in planting blocks
one and two were 4.4 and 7.3, 1.2 and 1.3 dS/m, and 60 and
59 percent, respectively.
Planting Stock
Seedlings of Apache plume (Questa, NM, seed source) and
serviceberry (Utah seed source) were propagated at the
Natural Resources Conservation Service Plant Materials
3
Center in Los Lunas, NM. Seedlings were grown in 164-cm
containers (Ray Leach Super Cells) in a peat:perlite growing
media (two parts peat moss and one part perlite by volume).
Plants were fertilized with a water soluble 20-10-20 fertilizer (Peter’s Peat Lite Special).
USDA Forest Service Proceedings RMRS-P-31. 2004
Williams, Tahboub, Harrington, Dreesen, Ulery
Fertilization Treatments
Prior to transplanting, sites were ripped to a depth of 45
cm and irrigated. Ripping was accomplished using three 65cm ripping bars attached to the back of a crawler tractor. In
August 1995, seedlings (approximately 10 to 20 cm tall) were
transplanted into the two sites in a randomized complete
block design and irrigated the day after planting. Three
blocks per site were established and each block had two
parallel rows, 50 cm apart. Within each row, plant spacing
was 30 cm. One row of each replicate block received fertilization treatment at the time of planting and the other row
did not receive fertilizer (control plots). Six grams of Sierra,
Inc., 17-6-12 plus micronutrients slow-release fertilizer
(Scotts Company, Marysville, OH) were placed into each
planting hole prior to transplanting the seedlings. Release
duration of this fertilizer is 3 to 4 months at 21 ∞C. From 1996
through 2000 all plants received supplemental fertilization
once each year.
Survival and Shoot Growth
In September 1996 and August 2000 survival of both
species was documented. In Spring 2001, shoot growth
(height and crown width) was measured to the nearest
centimeter for each plant. An average crown width was
calculated for each plant using two perpendicular measurements of crown width oriented at 45 degrees to the direction
of the planting row.
Root Measurements
In blocks one and two from each site, two plants per
species per fertilization treatment were measured for root
growth and distribution. Roots were evaluated using techniques described by Parsons and others (1998). Initial excavation in November 2000 was performed using a backhoe to
create a trench 1.5 m deep and 2 m long, 45 cm from the base
of each shrub row. A 30 by 30 cm vertical plane was hand
excavated 30 cm from the base of each plant where a 30 by
30 cm sampling frame was placed for root evaluation. The
frame was constructed out of clear Plexiglas and divided by
lines into 36, 5 by 5 cm grid cells. In each grid cell, roots were
counted and divided into three diameter classes: <0.5, 0.5 to
2.0, and >2.0 mm. Root measurements were repeated at 20
and 10 cm from the base of each plant within the same
vertical sampling frame.
Data Analysis
The experimental design was a completely randomized
design set in a split-split plot. Whole plot treatment was
site and block within site was the whole plot error term.
The split factor was fertilization treatment and the splitsplit factors were plane and depth. “Plane” refers to the
horizontal distance away from the plant perpendicular to
the row (10, 20, or 30 cm).
Survival of each species in 1996 and 2000 was analyzed
through Chi-square tests in SAS using PROC FREQ (SAS
Institute 1997). Shoot growth (height and crown width) was
analyzed for each species using analysis of variance in SAS
25
Root Growth of Apache Plume and Serviceberry on Molybdenum Mine Overburden...
(PROC GLM) to indicate simple and interaction effects of
site and treatment.
Root counts were analyzed separately for each species
using analysis of variance in a 2 (site) x 2 (fertilization
treatment) x 3 (plane) x 6 (depth) factorial (PROC MIXED,
SAS Institute 1997). Roots from three categories were combined for a weighted total to obtain a total root density
2
(number of roots per 150 cm ). Roots 0 to 0.05 mm were
assigned a midpoint value of 0.025, roots 0.05 to 2.00 mm
were assigned a midpoint value of 1.25, and roots greater
than 2.00 mm were assigned a midpoint value of 2.25.
For root response, PROC MIXED calculated F statistics,
means, and standard errors of both main effects and interaction combinations. Main effects of plane and depth and
their interactions with fertilization treatments were evaluated, and least significant differences (LSD) were carried
out for pairwise comparisons of main effects of plane and
depth on root growth. All treatment effects for survival, and
shoot and root growth were evaluated for significance at the
0.05 alpha level.
Results ________________________
6
Control
Total root density (# roots/150 cm2)
Williams, Tahboub, Harrington, Dreesen, Ulery
Fertilize
5
Mean effect
4
3
2
1
0
0-5
5-10
10-15
15-20
20-25
25-30
Overburden depth (cm)
Figure 1—Total root density of serviceberry for control
and fertilization treatments across overburden depths.
Mean represents the effect of overburden depth
averaged across fertilization treatments. Bars
represent ± one standard error, n = 4.
Saskatoon Serviceberry
5.0
Total root density (# roots/150 cm2)
Fertilization at time of planting increased serviceberry
crown width relative to unfertilized control plants (p =
0.0015). Crown widths of fertilized and control plants were
31.2 and 19.0 cm, respectively. However, fertilization at
time of planting reduced survival of serviceberry, compared
to controls, at both planting sites in 1996 (P = 0.0006) and
2000 (p = 0.0002). Survival of nonfertilized control plants
was highest at the Spring Gulch site (table 1).
Fertilization at time of planting influenced total root
density of serviceberry depending on sampling depth (P <
0.0001; fig. 1). Fertilized plants had lower total root densities at 5- to 10- and 10- to 15-cm depths than control plants.
Across horizontal planes, fertilization treatments influenced
total root density of serviceberry (P < 0.010, fig. 2). Serviceberry plants not fertilized at time of planting had greater
total root density in the 10-cm plane than fertilized plants.
Control
4.5
Fertilize
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
10
Apache Plume
Shoot growth of Apache plume plants fertilized at time
of planting was greater than unfertilized plants (height,
P < 0.0001; crown width, P = 0.0022). Average height and
Table 1—One- and 5-year survival of serviceberry by site and fertilization
at time of planting.
Site
Treatment
Spring Gulch
Blind Gulch
Spring Gulch
Blind Gulch
Control
Control
Fertilize
Fertilize
a
b
26
20
30
Distance (plane) (cm)
Survivala
1996
2000
15
10
6
5
Survival as in number of individual plants, n = 15.
Percent survival in 2000.
14
10
4
4
Percentb
93
67
27
27
Figure 2—Total root density of serviceberry for control
and fertilization treatments at three distances
(planes). Bars represent ± one standard error, n = 4.
crown width of fertilized plants were 42.4 and 43.2 cm, while
average height and crown width of control plants were 21.3
and 29.0 cm, respectively. Survival of Apache plume was
influenced by site and fertilization treatments in 1996 (p =
0.0035) and 2000 (p = 0.0018). Nonfertilized plants at Spring
Gulch had the highest survival in both years out of all site by
treatment combinations, while fertilized plants at Blind
Gulch had the lowest survival in 1996 and 2000 (table 2).
Fertilization at time of planting influenced Apache plume
total root density depending on depth and plane (P = 0.0012;
USDA Forest Service Proceedings RMRS-P-31. 2004
Root Growth of Apache Plume and Serviceberry on Molybdenum Mine Overburden...
Williams, Tahboub, Harrington, Dreesen, Ulery
Table 2 —One- and five-year survival of Apache plume by site and
fertilization at time of planting.
Site
Treatment
Spring Gulch
Blind Gulch
Spring Gulch
Blind Gulch
Control
Control
Fertilize
Fertilize
Survivala
1996
2000
21
13
16
11
20
12
16
9
10
(a) 10-cm plane
8
Percent
b
6
95
57
76
43
4
2
a
Survival as in number of individual plants, n = 21.
b
Percent survival in 2000.
Discussion _____________________
Survival and Shoot Growth
Survival of fertilized serviceberry and Apache plume
plants was low the first year after planting, although fertilized plants that survived were larger than control plants.
Increased shoot growth with fertilizer applications is concurrent with other literature. Burgess and others (1995)
found NPK fertilization treatments improved shoot growth
of white pine (Pinus strobiformis). Fertilizer applications on
a reclamation site in Idaho increased growth of both native
and introduced plant species (Williams and others 1990).
Nitrogen and phosphorus applications greatly increased
pine seedling height on mine overburden in Alabama (Zarger
and others 1973).
Low survival has been associated with the use of slowrelease fertilizers in other studies. Fisher and others (1983)
observed slow-release fertilizer caused high mortality in
juniper seedlings (Juniperus monosperma). At an eastern
Sierra Nevada surface mine, slow-release fertilizers applied
at doses of 30 g increased mortality of containerized, transplanted Jeffrey pine (Pinus jeffreyi), whereas doses of 10 and
20 g did not affect mortality (Walker 1999b). Transplanting
date and release time of the fertilizer used in the study may
have caused the observed mortality during the first year for
serviceberry. Fisher and others (1983) suggest transplanting and fertilizing containerized tree seedlings during the
summer rainfall period (July) in the Southwest rather than
in late summer and fall. Slow-release fertilizer applications
in August were more detrimental to juniper seedling growth
and survival than May and July applications. It was speculated that low night temperatures following recent nutrient
additions resulted in frost damage in August (Fisher and
others 1983). Mortality of ponderosa pine seedlings in northern Idaho was influenced by dose and rate of slow-release
fertilizer, with high doses and 9- and 12-month release times
increasing mortality (Fan and others 2002).
Survival and shoot growth of transplanted and fertilized
woody seedlings appear to be influenced by many abiotic
factors including planting date and the amount, formulation, and rate of release of slow-release fertilizer. Based on
published reports and our findings, transplanting Apache
plume and serviceberry earlier in the growing season, early
USDA Forest Service Proceedings RMRS-P-31. 2004
Total root density (# roots/150 cm2)
fig. 3). In the 20-cm plane, total root density of fertilized
plants was greater at depths below 20 cm than all other
treatments.
0
-2
10
(b) 20-cm plane
8
6
4
2
0
-2
10
(c) 30-cm plane
8
6
4
2
0
-2
0-5
5-10
10-15
15-20
20-25
25-30
Overburden depth (cm)
Control
Fertilize
Figure 3—Total root density of Apache plume for
control and fertilization treatments across
overburden depths at three distances (planes).
Bars represent ± one standard error, n = 4.
to midsummer, with a concurrent application of a slowrelease fertilizer that releases nutrients when plants are
actively growing, may reduce mortality and promote shoot
growth for both species.
Root Growth
Fertilization treatment effects on plant root density and
distribution vary in the literature. In this study, there was
no main effect of fertilization on overall root density for
27
Williams, Tahboub, Harrington, Dreesen, Ulery
either species, but fertilization at time of planting did affect
root distribution. Studies have shown alteration in plant
root distribution in relation to fertilizer applications. For
example, grasses increased their root growth in response to
increased soil nutrients (Eissenstat and Caldwell 1988).
Friend and others (1990) found that Douglas-fir plants
produced more roots in nitrogen-rich than in nitrogen-poor
microenvironments. In nitrogen stressed environments,
Douglas-fir plants had a greater frequency of roots within
nitrogen rich microenvironments than in nonstressed environments (Friend and others 1990). Similarly, Ringwall and
others (2000) observed that fertilized leafy spurge plants
allocated a greater portion of root biomass within the first 10
cm of soil and distributed a larger portion of roots in fertilized areas of the planting medium. It appears that root
system distribution can be manipulated by fertilization at
time of planting, but responses may be species dependent, as
seen in our study.
Fertilization at time of planting had a negative effect on
total root density of serviceberry at the base of the plant
relative to the nonfertilized control. The rapid release of
nutrients coupled with transplanting in late summer when
plants were still actively growing may have promoted shoot
development at the expense of root development into the
overburden. As a consequence, plants receiving fertilizer at
time of planting may have had insufficient root systems to
support their growth and survival during the first year
following plantings. Fertilization had an overall positive
effect on total root density of Apache plume deeper in the
profile at 20 cm from the base of the plant. Although
fertilization decreased survival of Apache plume, its effect
on shoot growth and root density suggests that the treatments were not as detrimental to the plant’s performance as
it was in serviceberry. Species were analyzed separately in
this study; however, comparisons as to their response to
fertilization at time of planting are important for future
decisions about species selection and revegetation designs.
Fertilizer applications at time of planting may increase
the establishment of herbaceous species (annuals, grasses,
and forbs), which can negatively impact (for example, nutrient and water competition) survival and shoot and root
growth of transplanted shrubs and trees (Cook and others
1974). Although not measured, plots fertilized at time of
planting did have an observable increase of volunteer herbaceous plants (grasses and annual forbs) compared to control
plots in 2000. It is not known when these plants became
established, but favorable conditions did exist for them to
establish within these plots. Favorable conditions may include fertilizer applications and lack of root competition
near the soil surface (<20 cm), since both species fertilized at
time of planting allocated a greater portion of roots deeper in
the soil than at the surface. However, further evaluations
are needed to determine whether the observed root response
in both species was due directly to fertilizer applications or
indirectly from competition from volunteer plants, or both.
Conclusions ____________________
Although overall performance in terms of shoot growth
was positive for both species when fertilized at time of
28
Root Growth of Apache Plume and Serviceberry on Molybdenum Mine Overburden...
planting, other fertilizer-related factors will need to be
investigated to enhance establishment, growth, and longterm survival of plants planted directly into mine overburden at high elevations. Slow-release fertilizer formulation,
release rate and total amount applied should be investigated
more closely. While these factors have been investigated in
other systems (Fan and others 2002; Walker 1999a,b), the
impact of these treatments on plants at high elevations
(>2,500 m) remains largely unknown.
Additional factors, which need further investigation, are
related to fertilizer treatments for high elevation planting
and include time of planting relative to the end of the
growing season. Previous studies at this mine indicate two
periods suitable, in terms of available moisture, for transplanting. These are during the midsummer rain period and
in early fall. Incorporation of fertilizer, in particular nitrogen fertilizer, on fall plantings is not recommended; however, further work is warranted to look at planting and
fertilizing earlier in the growing season than was performed in this study. The effects of incorporating slowrelease fertilizer at this time will impact plant response.
Too slow a release rate or too short a timeframe to the end
of the growing season may increase mortality rates. Transplanting in midsummer, as recommended by Fisher and
others (1983), may have prevented mortality in both species when fertilized. Timed properly this treatment can
result in larger, more vigorous plants.
References _____________________
Allen, M. F. 1989. Mycorrhizae and rehabilitation of disturbed arid
soils: processes and practices. Arid Soil Research and Rehabilitation. 3: 229–241.
Boyce, Richard L.; Lucero, Scott A. 1999. Role of roots in winter
water relations of Engelmann spruce saplings. Tree Physiology.
19: 893–898.
Brown, R. W.; Amacher, M. C.; Williams, B. D.; Mueggler, W. F.;
Kotuby-Amacher, J. 1996. Revegetation of acidic mine spoils at
high elevations: restoration of native plant communities. Association of Abandoned Mine Land Programs Conference; 1996
September 15–18; Kalispell, MT.
Burgess, D.; Baldock, J. A.; Wetzell, S.; Brand, D. G. 1995. Scarification, fertilization and herbicide treatment effects on planted
conifers and soil fertility. Plant and Soil. 169: 513–522.
Cook, C. W.; Hyde, R. M.; Sims, P. L. 1974. Revegetation guidelines
for surface mined areas. Science Series No 16. Fort Collins, CO:
Colorado State University, Range Science Department.
Eissenstat, D. M.; Caldwell, M. M. 1988. Seasonal timing of root
growth in favorable microsites. Ecology. 69: 870–873.
Fan, Zhaofei; Moore, James A.; Shafii, Bahman; Osborne, Harold L.
2002. Three-year response of ponderosa pine seedlings to controlled-release fertilizer applied at planting. Western Journal of
Applied Forestry. 17: 154–164.
Feagley, S. E. 1985. Chemical, physical and mineralogical characteristics of the overburden. Southern Coop Series Bull. 294.
Fayetteville: Arkansas Agricultural Experimental Station.
Fisher, James T.; McRae, John B.; Aldon, Earl F. 1983. Methods for
establishing containerized native juniper on surface disturbed
sites in the Southwest. In: Pope, P. E., ed. Better reclamation with
trees conference: proceedings; 1983 June 2–3; Terre Haute, IN.
Purdue University: 76–88.
Friend, Alexander L.; Eide, Marvin R.; Hinckley, Thomas M. 1990.
Nitrogen stress alter root proliferation in Douglas-fir seedlings.
Canadian Journal of Forest Research. 20: 1524–1529.
Gardiner, D. T. 1993. Revegetation status of reclaimed abandoned
mined land in western North Dakota. Arid Soil Research and
Rehabilitation. 7: 79–84.
USDA Forest Service Proceedings RMRS-P-31. 2004
Root Growth of Apache Plume and Serviceberry on Molybdenum Mine Overburden...
Gleason, J. F.; Duryea, M.; Rose, R.; Atkinson, M. 1990. Nursery and
fertilization of 2+0 ponderosa pine seedlings: the effect on morphology, physiology, and field performance. Canadian Journal of
Forest Research. 20: 1766–1772.
Harrington, J. T.; Dreesen, D. R.; Wagner, A. M.; Murray, L.; Sun, P.
2001a. Results of species trials on low pH overburden materials for
mine land reclamation. In: Barnhisel, R. I.; Buchanan, B. A.;
Peterson, D.; Pfeil, J. J., comps. Land reclamation: a different
approach: proceedings; 2001 June 3–7; Albuquerque, NM. Lexington, KY: American Society of Surface Mining and Reclamation.
Harrington, J. T.; Dreesen, D. R.; Wagner, A. M.; Murray, L.; Sun, P.
2001b. The influence of seed source and stock size on first-year
performance of direct transplanted conifer seedlings. In: Barnhisel,
R. I.; Buchanan, B. A.; Peterson, D.; Pfeil, J. J., comps. Land
reclamation: a different approach: proceedings; 2001 June 3–7;
Albuquerque, NM. Lexington, KY: American Society of Surface
Mining and Reclamation.
Harrington, J. T.; Wagner, A. M.; Dreesen, D. R. 2001c. Influence of
pisolithus tinctorius inoculation on greenhouse growth and firstyear transplant survival of conifer seedlings. In: Barnhisel, R. I.;
Buchanan, B. A.; Peterson, D.; Pfeil, J. J., comps. Land reclamation: a different approach: proceedings; 2001 June 3–7; Albuquerque, NM. Lexington, KY: American Society of Surface Mining and
Reclamation.
Houle, G.; Babeux, P. 1994. Fertilizing and mulching influence on
the performance of four native woody species suitable for revegetation in subarctic Quebec. Canadian Journal of Forest Research. 24: 2342–2349.
Meyer, J.; Leonardson, R. 1990. Tectonic, hydrothermal and geomorphic controls on alteration scar formation near Questa, New
Mexico. In: New Mexico Geologic Society guidebook 41st field
conference; Southern Sangre de Cristo Mountains, New Mexico:
417–422.
Parsons, W. F.; Ehrenfeld, J. G.; Handel, S. N. 1998. Vertical growth
and mycorrhizal infection of woody plant roots as potential limits
to the restoration of woodlands on landfills. Restoration Ecology.
6(3): 280–289.
Ringwall, Kirstin; Biondini, Mario E.; Grygiel, Carolyn E. 2000.
Effects of nitrogen fertilization in leafy spurge root architecture.
Journal of Range Management. 53: 228–232.
USDA Forest Service Proceedings RMRS-P-31. 2004
Williams, Tahboub, Harrington, Dreesen, Ulery
SAS Institute Inc. 1997. SAS/STAT® software: changes and enhancements through release 6.12. Cary, NC: SAS Institute Inc.
1167 p.
Schilling, J. H. 1965. Molybdenum resources of New Mexico. Socorro,
NM: State Bureau of Mines and Minerals Resources, New Mexico
Institute of Mining and Technology Campus Station. 76 p.
State of New Mexico. 1999. New Mexico Natural Resources: data
and statistics for 1998. New Mexico Energy, Minerals and Natural Resources Department. Santa Fe: New Mexico Printing Office. 65 p.
Steffen, Robertson; Kirsten (SRK Inc.) 1995. Questa molybdenum
mine geochemical assessment: Molycorp Inc. Lakewood, CO.
Voeller, P. J.; Zamora, B. A.; Harsh, J. 1998. Growth response of
native shrubs to acid mine spoil and to proposed soil amendments. Plant and Soil. 198: 209–217.
Wagner, A. M.; Harrington, J. T. 1994. Revegetation report for
Molycorp, Inc. New Mexico Mine Permit TA001RE. Questa, NM.
Walker, R. F. 1999a. Artificial regeneration of Jeffrey pine in the
Sierra Nevada: growth, nutrition, and water relations as influenced by controlled release fertilization and solar protection.
Journal of Sustainable Forestry. 9: 23–39.
Walker, R. F. 1999b. Reforestation of an eastern Sierra Nevada
surface mine with containerized Jeffrey pine: seedling growth
and nutritional responses to controlled release fertilization and
ectomycorrhizal inoculation. Journal of Sustainable Forestry. 9:
127–147.
Williams, B. D.; Brown, R. W.; Sidle, R. C.; Mueggler, W. F. 1990.
Greenhouse evaluation of reclamation treatments for perlitepumice mine spoils. Res. Pap. INT-426. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research
Station. 7 p.
Zarger, T. G.; Bengston, G. W.; Allen, J. C.; Mays, D. A. 1973. Use
of fertilizers to speed pine establishment on reclaimed coal-mine
spoil in northeastern Alabama: II. Field experiments. In: Hutnik,
R. J.; Davis, R., ed. Ecology and reclamation of devastated land:
volume II. New York: Gorden and Breach.
29
Importance of Genotype, Soil
Type, and Location on the
Performance of Parental and
Hybrid Big Sagebrush Reciprocal
Transplants in the Gardens of
Salt Creek Canyon
Kathleen J. Miglia
D. Carl Freeman
E. Durant McArthur
Bruce N. Smith
Abstract: The majority of flowering plant species are descendent
from hybridization events; therefore, understanding hybridization
in nature is essential to understanding plant speciation. Hybrid
zones, common and often stable, provide one source of hybridization. Stable hybrid zones are particularly important in evolutionary
theory because they violate adaptive speciation theory and call into
question the universality of reproductive isolation in speciation.
Under this model, hybrids are assumed to be unfit regardless of
environment, due to endogenous selection, thereby contributing to
reproductive isolation. However, a number of stable hybrid zones
show hybrids to be most fit within the hybrid zone due to ecological
selection. Regardless of the type of selection occurring, stability of
the hybrid zone results from a balance between gene flow and
selection. What is not known is how hybridization allows hybrids to
become adapted to the habitat of the hybrid zone. An understanding
of the interactions between novel hybrid plant genotypes and their
ecological habitats is important to understanding plant speciation.
Results from a reciprocal transplant experiment involving the big
sagebrush hybrid zone (Artemisia tridentata ssp. tridentata x A. t.
ssp. vaseyana) in Salt Creek Canyon, UT, show that the parental
and hybrid genotypes are not adapted to either the soils or location.
Rather, it appears that the microorganisms in the soils are adapted
to their location. It is possible that the plants are adapted to the
microorganisms in their native soils. Our data also do not support
the common assumption of hybrid unfitness.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Kathleen J. Miglia is Research Associate and D. Carl Freeman is Professor,
Department of Biological Sciences, Wayne State University, Detroit, MI
48202, U.S.A.; e-mail: kmiglia@sun.science.wayne.edu. E. Durant McArthur
is Project Leader, Shrub Sciences Laboratory, Rocky Mountain Research
Station, USDA Forest Service, Provo, UT 84606, U.S.A. Bruce N. Smith is
Professor, Department of Plant and Animal Sciences, Brigham Young University, Provo, UT 84602, U.S.A.
30
Introduction ____________________
Hybridization plays a key role in adaptive speciation
theory (Dobzhansky 1940; Grant 1963; Mayr 1942; Stebbins
1950) because it is necessary for reinforcement of premating
isolating barriers. Accordingly, the parental taxa are believed to diverge genetically due to some period of isolation,
and subsequently come into contact again forming hybrids.
If these hybrids are unfit they will be selected against, and
genes that cause the parental species not to mate with each
other will spread, generating or reinforcing premating isolating barriers, with the hybrids and hybridization eliminated over time. However, if hybrids are more fit than the
parental taxa, then introgression will occur with the replacement of the parental taxa by introgressed populations.
Either way the hybrid zone should be ephemeral. Stable
hybrid zones violate adaptive speciation theory because
hybrids can persist indefinitely (for example, the cottonwood
hybrid zone studied by Eckenwalder [1984] is believed to be
2 million years old).
Three models (reviewed by Arnold 1997) seek to explain
stable hybrid zones, with each postulating that selection
balances gene flow. The models differ in the nature of that
selection and the importance of the environment. Stable
plant hybrid zones are particularly useful for studying these
models because plants can be easily used in reciprocal
transplant experiments (Emms and Arnold 1997; Wang and
others 1997) and are often well adapted to local environmental conditions (reviewed in Linhart and Grant 1997), which
is a key feature in adaptive speciation.
The Dynamic Equilibrium Model (Barton 1979a,b; Barton
and Hewitt 1985, 1989; Hewitt 1988) posits that endogenous
selection against hybrids balances gene flow into the hybrid
zone, thereby accounting for hybrid zone stability. It is
predicted that the hybrid zone will become trapped in areas
of low population density, fixing it in space, but otherwise
the model is environmentally neutral. Data from the big
sagebrush hybrid zone (the subject of this report) does not
support this model for two main reasons: big sagebrush
USDA Forest Service Proceedings RMRS-P-31. 2004
Importance of Genotype, Soil Type, and Location on the Performance ...
hybrids are more fit than the parental taxa within the
boundaries of the hybrid zone, but less fit outside it (Wang
and others 1997), and the hybrid zone does not occur in a
population density trough (Freeman and others 1999a,b).
The Mosaic Hybrid Zone Model (Britch and others 2001;
Harrison 1990; Harrison and Rand 1989; Howard and others
1993; Rand and Harrison 1989; Rieseberg and others 1998;
Ross and Harrison 2002) also posits endogenous selection
against hybrids, but assumes that the hybrid zone is a
mosaic of parental habitats. Exogenous selection acts against
the nonindigenous parental genotype, and while hybrids are
always produced, their formation is balanced by endogenous
selection against them. Again, data from the parental and
hybrid big sagebrush at Salt Creek Canyon do not support
this model because hybrids are the most fit genotype within
the hybrid zone (that is, there is no endogenous selection
against hybrids in the hybrid zone) and the hybrid zone is not
a patchwork of the two parental habitat types (Freeman and
others 1999a; Wang and others 1997, 1998).
The Bounded Hybrid Superiority Model (Moore 1977)
assumes hybrid superiority in the hybrid zone, with selection against nonindigenous genotypes in both the parental
and hybrid zone habitats. Hybrids are thus believed to be
adapted within, but not outside, the hybrid zone, with the
parental genotypes likewise superiorly adapted to their
indigenous habitats. Prior results of our reciprocal transplant experiments are entirely consistent with this model
(Freeman and others 1999a; Wang and others 1997, 1998,
1999). Implicit in this theory is that the parental and hybrid
habitats differ from one another, as do the niches of the
parental taxa and their hybrids. Wang and others (1998,
1999) found such niche separation when they analyzed the
elemental composition of soils and leaves from plants in the
reciprocal transplant gardens at Salt Creek Canyon. For
example, basin habitat soils are deeper, and have a higher
pH and lower concentrations of K, Mg, and Ba than soils
from the mountain habitat. The soils in the hybrid zone have
some elements in greater (Ca, K, Na) or lesser (Cu) concentrations than the soils of either parental habitat; therefore,
they cannot be simply considered intermediate between the
parental populations (Wang and others 1998). Similarly, the
concentration of B in the leaves of basin parental plants was
three times higher in the basin garden than in either the
middle hybrid or mountain gardens, while the concentration
of B in leaves of middle hybrid plants was nearly three times
greater in the middle hybrid zone garden than in either the
basin or mountain gardens. Mountain plants exhibited equal
concentrations of B in their leaves in the two parental
gardens, but markedly lower concentrations in the middle
hybrid zone garden, and the concentration of B in their
leaves was markedly lower overall compared to the other
genotypes. In conclusion, the genotype by environment interactions for leaf elemental concentrations of plants all
grown in the same three gardens strongly indicated niche
separation.
Whittam (1989) proposed that herbivorous insects are
more abundant in plant hybrid zones because hybridization
disrupts defensive coadapted gene complexes, rendering
hybrids more palatable than parental populations. We examined phytophagous insects in the Salt Creek reciprocal
transplant gardens (Graham and others 2001a,b). The hybrid
USDA Forest Service Proceedings RMRS-P-31. 2004
Miglia, Freeman, McArthur, and Smith
zone and each parental habitat have unique insect communities, but the hybrids did not suffer higher insect densities.
This again illustrates the different niches occupied by the
different genotypes.
Plant community structure was not examined across the
hybrid zone at Salt Creek Canyon because the area was
reseeded; consequently, we examined the plant communities in a big sagebrush hybrid zone at Clear Creek Canyon in
South Central Utah. We found that less than one-third of the
species occur in both parental habitats; one-fourth of the
species are unique to the basin habitat, and over half are
unique to the mountain habitat. Fifteen species occur only
within the hybrid zone (Freeman and others 1999a).
Together our prior data show that (1) basin and mountain
big sagebrush and their hybrids are components of different
biotic communities, (2) the hybrid zone occurs at an ecotone,
and (3) hybrids are not universally unfit as the Dynamic
Equilibrium and Mosaic Hybrid Zone models predict. Instead, genotype by environment interactions appear to be
stabilizing the big sagebrush hybrid zone, in keeping with
the predictions of the Bounded Hybrid Superiority model.
Methods _______________________
Our previous reciprocal transplant experiments involved
growing parental and hybrid seedlings in the basin, mountain, and middle hybrid zone habitats. Because the taxa are
parapatrically distributed along an elevational gradient,
these earlier transplant experiments confounded both location and soils; consequently, we were unable to sort out the
nature of the adaptation of the parental taxa and hybrids to
their respective indigenous habitats. This confounding is
especially true between the soils and temperature. Here, we
report on a reciprocal transplant experiment in which both
the soils and big sagebrush plants were transplanted into
the gardens at Salt Creek Canyon.
Big Sagebrush
The big sagebrush (Artemisia tridentata) complex is a
group of long-lived perennial, evergreen shrubs that dominates the landscape in the Western United States, providing
forage for livestock and wildlife (Beetle 1960; McArthur
1994; Trimble 1989; Wambolt 1996; Welch and McArthur
1986). The species is composed of five subspecies: A. t. ssp.
tridentata (basin big sagebrush), A. t. ssp. vaseyana (mountain big sagebrush), A. t. ssp. wyomingensis (Wyoming big
sagebrush), A. t. ssp. spiciformis (snowbank big sagebrush),
and A. t. ssp. xericensis (xeric big sagebrush). We examined
one hybrid zone between basin and mountain big sagebrush.
These subspecies differ genetically, morphologically, and
ecologically from each other (table 1) and are sympatrically
or parapatrically distributed, forming narrow hybrid zones
along the sides of mountains or other points of contact.
Study Site
The study site is located in Salt Creek Canyon, near
Nephi, in Juab County, UT, where basin and mountain big
sagebrush are parapatrically distributed, with the basin
subspecies occurring below 1,790 m in elevation and the
31
Miglia, Freeman, McArthur, and Smith
Importance of Genotype, Soil Type, and Location on the Performance ...
Table 1—Some characteristics that differ between basin and mountain big sagebrush.
Characteristic
Mountain
Basin
Reference
Habitat
Shallow, well-drained soils on
foothills and mountains
Dry, deep, well-drained alluvial soils
on plains, valleys, and foothills
McArthur 1994
Height
0.7 to 1.2 m
1.0 to 4.0 m
McArthur and Plummer 1978
Root system
Shallow
Deep
Welch and Jacobson 1988
Shoot morphology
• Spreading branches with an
even-topped crown
• Main stem is usually divided
• Branches are layered
• Erect, heavily branched shrub with
an uneven-topped crown
• Main stem is undivided (trunk-like)
• Branches are not layered
McArthur and others 1979
Leaf morphology
Broadly cuneate
Narrowly lanceolate
Beetle and Young 1965
Inflorescence
Spikate with many heads
Paniculate with few heads
Beetle and Young 1965
Palatability to mule deer
More palatable
Less palatable
Welch and McArthur 1986
mountain subspecies at elevations ranging from 1,850 m to
timberline on neighboring Mt. Nebo (Graham and others
1995). The hybrid zone is a narrow band, approximately
380 m wide (per measurements by EDM and KJM in 2001),
situated between the parental populations at elevations
ranging from 1,790 to 1,830 m. Three fenced common
gardens (8 by 15 m) were established across the hybrid
zone in October 1994: one in each of the parental populations and one in the middle of the hybrid zone (Wang 1996).
Experimental Procedure
Battery operated HOBO™ Weather Station Data Loggers
(Onset Computer Corp., Bourne, MA) were installed at each
garden. This type of weather station can store up to 500,000
measurements, even if the batteries fail. Hourly temperatures were measured from September 12, 2000, until July
10, 2001. Hourly and daily average temperatures were
analyzed using a mixed model ANOVA with site crossed
with month and day nested within month. Daily, maximum,
and minimum recorded temperatures were analyzed using
a factorial design.
Putative reciprocal F1, parental and indigenous hybrid
seeds (B x M, M x B, B x B, M x M, and H x H, respectively,
where B = basin big sagebrush, M = mountain big sagebrush, and H = hybrid between basin and mountain big
sagebrush) were made in the field following the procedures
of McArthur and others (1988). The first letter denotes the
maternal parent and the second the paternal parent. While
pollinations were highly controlled, some self-pollination
may have occurred. In May 1999, seeds were germinated
and subsequent seedlings raised in randomly arranged
pots in the USDA Forest Service Shrub Sciences Laboratory greenhouse in Provo, UT. Seedlings were given equal
amounts of water (table 2) when needed and pots rotated
approximately every 6 weeks to balance any effects due to
uneven greenhouse conditions. Greenhouse temperatures
were controlled for favorable growth conditions (table 3). A
15:30:15 (available, water-soluble nitrogen:phosphate:
potash) fertilizer solution was applied twice to the pots on
January 25, 2000, and March 13, 2000. Five replicates of
each genotype were planted in each of three soil types
32
(basin, mountain, and middle-hybrid zone) collected at
sites immediately adjacent to the three common gardens.
Three trenches (approximately 60 cm wide, 90 cm deep,
and 12 m long) were excavated with a backhoe in each
garden in May 2000, lined with 4 ml polyethylene sheeting,
Table 2—The watering schedule of seedlings during their stay in the
greenhouse. The amounts listed are approximate averages.
Date
Amount
Date
Amount
6/24/99
6/26/99
6/28/99
7/01/99
7/06/99
7/08/99
7/12/99
7/14/99
7/16/99
7/19/99
7/21/99
7/23/99
7/26/99
7/28/99
7/30/99
8/01/99
8/04/99
8/07/99
8/09/99
8/12/99
8/14/99
8/17/99
8/20/99
8/23/99
8/26/99
8/29/99
9/01/99
9/07/99
9/09/99
9/11/99
9/14/99
9/17/99
mL
104.9
61.8
193.1
84.5
118.5
106.5
87.7
117.6
77.9
122.8
53.8
120.9
94.7
118.8
83.6
115.3
102.4
74.2
128.7
123.4
99.8
102.6
121.0
101.0
70.6
57.8
103.9
120.1
100.5
96.4
114
150.1
9/20/99
9/24/99
9/28/99
10/01/99
10/05/99
10/09/99
10/13/99
10/15/99
10/21/99
10/27/99
11/01/99
11/04/99
11/08/99
11/12/99
11/16/99
11/22/99
11/27/99
12/01/99
12/06/99
12/13/99
12/16/99
12/20/99
12/23/99
12/26/99
12/29/99
1/02/00
1/06/00
1/10/00
1/14/00
1/17/00
1/21/00
1/25/00
mL
136.3
86.5
91
111.7
82.2
124.4
104.7
97.9
72.8
111.1
90.0
111.4
98.8
100.0
136.5
85.5
160.7
78.4
125.1
87.5
186.6
116.0
70.8
60.1
80.6
108.6
75.3
66.8
115.7
75.3
87.3
86.0
Date
Amount
1/28/00
2/01/00
2/04/00
2/07/00
2/11/00
2/15/00
2/18/00
2/22/00
2/25/00
2/28/00
3/03/00
3/06/00
3/10/00
3/13/00
3/17/00
3/20/00
3/24/00
3/27/00
3/31/00
4/03/00
4/06/00
4/09/00
4/12/00
4/14/00
4/17/00
4/19/00
4/21/00
4/24/00
4/28/00
5/01/00
mL
93.1
129.7
101.3
115.8
88.0
109.8
74.8
86.4
79.4
87.5
106.7
93.7
117.2
157.4
160.7
164.2
137.5
152.9
166.1
137.9
160.5
135.5
187.9
207.8
187.6
190.8
69.0
170.0
174.0
125.0
USDA Forest Service Proceedings RMRS-P-31. 2004
Importance of Genotype, Soil Type, and Location on the Performance ...
Miglia, Freeman, McArthur, and Smith
Table 3—The average minimum and maximum temperatures in the
greenhouse during the first year of seedling growth.
Month/year
6/99
7/99
8/99
9/99
10/99
11/99
12/99
1/00
2/00
3/00
4/00
5/00
Minimum
temperature
Maximum
temperature
N
- - - - - - - - - - - -∞C - - - - - - - - - 13
41
15
40
16
38
9
35
8
39
8
35
9
34
9
31
9
34
10
37
10
39
11
36
21
31
31
30
31
30
31
31
29
31
30
8
and then filled with one of the same three soil types used in
the greenhouse (that is, each of the three soil types was
represented in each garden, including the indigenous soil
type for each garden). One-year-old seedlings were then
planted at random in the soil-filled trenches, with the
choice of trench corresponding to the soil type to which a
seedling was exposed during its stay in the greenhouse.
There were 25 transplants per trench, giving a total of 75
transplants per garden. Each plant was marked with a
labeled rebar. Plants were watered in the gardens weekly
until watering was tapered to once every other week in midAugust 2000 and then stopped completely in mid-September 2000.
Data Collection and Statistical Analyses
Measurements were taken in August 2001 and August
2002 for the following growth and reproductive parameters:
crown diameters 1 and 2 (crown 1 = greatest diameter of
plant; crown 2 = diameter directly perpendicular to crown 1),
tallest vegetative branch, tallest branch (tallest vegetative
branch plus tallest inflorescence branch, when present), and
length of three representative inflorescences
All temperature and morphological data were analyzed
using the SPSS version 10.0 statistical software (SPSS Inc.
2001). All morphological data were analyzed using a
MANOVA (SPSS Inc. 2001). Cases involving equal variances were followed by Bonferroni post hoc tests (SPSS).
Data violating the assumption of equal variances were
square-root transformed. In those cases where data transformation did not yield equal variances, Dunnett’s T3 post
hoc tests were performed (SPSS Inc. 2001 ).
daily minimum temperatures than either the basin or middle
hybrid zone garden sites (table 4). The average daily and
maximum temperatures did not differ significantly among
the gardens; however, the middle hybrid zone garden site
had both the lowest average daily temperature and highest
average maximum temperature (table 4).
Height
We measured height two ways. First, we measured the
height of the tallest vegetative branch and second, the
height of the tallest branch that included an inflorescence.
Both measures of height differed significantly among the
gardens (F2,152 = 8.117, and 2.54, P < 0.001 and 0.08, respectively). Plants in the middle hybrid zone garden were significantly taller than those in the parental gardens; the average
height of plants in the parental gardens did not differ from
each other (table 5). Vegetative height also differed significantly among the genotypes (F2,152 = 5.39, P < 0.001). Both
types of F1 hybrids were significantly shorter than either the
basin or indigenous hybrid plants. The F1 plants did not
differ in size from the mountain plants nor did mountain
plants differ from either the basin or indigenous hybrid
plants.
Vegetative height was independent of soil type; however,
when the inflorescences were included, the middle hybrid
zone soils produced significantly shorter plants than the
other two soil types, which did not differ from each other.
There was a highly significant garden by soil type interaction for both the vegetative and vegetative plus inflorescence measures of height (F4,152 = 3.035, and 7.574, respectively; P < 0.02 and 0.001, respectively). The trend was
most dramatic for the inflorescence measure. In the basin
garden, plants grew best in the basin soils. However, in the
middle hybrid zone garden, middle hybrid zone soils yielded
the greatest height. There was no trend observed for this
measure in the mountain garden (figs. 1b and 1c).
Crown Diameter
Average crown diameter differed significantly among the
genotypes (F4, 152 = 4.23, P < 0.003), but not among the
gardens or soil types (table 6). Basin plants produced significantly smaller crown diameters than the other four genotypes, which did not differ among themselves. The garden by
soil interaction was also significant (F4,152 = 3.45, P < 0.01)
(fig. 1d). Indigenous hybrids produced their greatest crown
diameter in the middle hybrid zone garden in the middle
hybrid zone soils, whereas both of the parental genotypes
Table 4—Temperature profiles in each of the three gardensa.
Results ________________________
Temperature
The average daily minimum temperature differed significantly among the sites (F2,873 = 19.20, P < 0.001), with the
basin garden site having the lowest average daily minimum
temperature and the mountain garden site having higher
USDA Forest Service Proceedings RMRS-P-31. 2004
Site
Average daily
temperature
Average daily
minimum
temperature
Average Daily
maximum
temperature
Basin
Hybrid
Mountain
8.10a
7.78a
8.14a
–2.01a
–1.64a
–0.10b
18.66a
19.68a
18.79a
a
Different letters in columns indicate significant (P < 0.05) differences among
values.
33
Miglia, Freeman, McArthur, and Smith
Importance of Genotype, Soil Type, and Location on the Performance ...
Table 5—Averages for vegetative height, vegetative plus inflorescence height, average inflorescence length and
crown diameter for plants in each of the three gardensa.
Garden
Basin
Hybrid
Mountain
a
Vegetative height
Height including
inflorescence
Average inflorescence
length
Average crown
diameter
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - cm - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 45.50a
48.62a
16.35a
43.56a
51.17b
54.22b
19.75b
48.37a
43.09a
44.94a
16.58a
43.99a
Different letters in columns indicate significant (P < 0.05) differences among values.
Table 6—Averages for vegetative height, vegetative plus inflorescence height, average inflorescence length and
crown diameter for each of the five genotypesa.
Genotype
BxB
BxM
MxB
MxM
HxH
a
Vegetative height
Height including
inflorescence
Average crown
diameter
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - cm - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 51.02a
46.37a
15.42a
38.54a
40.38b
41.41a
15.43a
48.02b
43.50b
52.80a
19.49a
45.39b
50.42a
54.02a
19.46a
46.68b
45.81ab
49.79a
17.52a
49.04b
Different letters in columns indicate significant (P < 0.05) differences among values.
produced their greatest crown diameters in the basin garden
in the basin soils.
Inflorescence Length
The average inflorescence length differed significantly
among the gardens F 2,152 = 2.87, P< 0.06) and soils (F2,152 =
5.74, P < 0.005), but not among the genotypes. Plants in the
middle hybrid zone garden produced significantly longer
inflorescences than in the other two gardens, which did not
differ from each other. Plants in the basin soils produced
longer inflorescences than did plants in the middle hybrid
zone soils. Inflorescence length of plants in the mountain
soils was intermediate between those in the middle hybrid
zone and basin soils and did not differ significantly from
either. There was a highly significant garden by soil interaction (F4,152 = 9.85, P < 0.001). Plants in the middle
hybrid zone soils performed very poorly in the basin garden, while in the middle hybrid zone garden, plants in the
middle hybrid zone soils performed better than in the other
two soil types (fig. 1a).
Discussion _____________________
Our findings indicate that the minimum temperature
differed greatly across the hybrid zone. This finding coupled
with those examining soils (Wang and others 1998), elemental leaf concentrations (Wang and others 1999), and the
insect (Graham and others 2001a,b) and plant communities
(Freeman and others 1999b) strongly indicate that the
habitats occupied by the parental taxa and their hybrids are
distinct from one another. Thus, the big sagebrush hybrid
zone occurs at an ecotone. Our results also confirm earlier
34
Average inflorescence
length
studies showing that the different subspecies of big sagebrush exhibit different morphologies and that these differences are preserved in common gardens—indicating a genetic basis for both the growth habits and niche separation
(McArthur 1994; McArthur and others 1979; McArthur and
Welch 1982). However, our most important result from this
study is somewhat disconcerting. Earlier work (Wang and
others 1997) had shown that basin and mountain big sagebrush are each adapted to their indigenous habitat and that
indigenous hybrids were the most fit genotype within the
hybrid zone. The transplant experiments involved in that
study confounded the physical location with the soils. In the
present study, we reciprocally transplanted both the soils
and seedlings into the gardens to sort out these effects, if
any. We anticipated that the seedlings might be adapted to
their indigenous soils and perhaps their indigenous physical
location, as well. This is not what we observed. Neither the
genotype-by-garden nor garden-by-soil interactions were
significant for any variable. However, the garden-by-soil
interaction was significant for every variable, which was
indicative of adaptation in two cases: basin soils in the basin
garden and middle hybrid zone soils in the middle hybrid
zone garden. This leads us to conclude that the indigenous
soils are adapted at each of these locations, implying that it
is the soil microorganisms that have become adapted to a
particular location and not the plants themselves. The
results of this study coupled with our earlier results (Wang
and others 1997) imply that the different genotypes must be
adapted to the indigenous microorganisms and that these
microorganisms are adapted to the soils and physical location in the garden of their indigenous habitat. We have yet
to determine whether or not this is case, and also if the
microorganisms in the middle of the hybrid zone are distinct
from those in either parental habitat. If our conclusions are
USDA Forest Service Proceedings RMRS-P-31. 2004
Importance of Genotype, Soil Type, and Location on the Performance ...
correct, then hybrid zone theory needs to be expanded to
include symbiotic interactions as well as the physical and
chemical aspects of the environment.
30
Average inflorescence length (cm)
Miglia, Freeman, McArthur, and Smith
A
20
Acknowledgments ______________
10
Soil
Basin
Hybrid
We express appreciation to Gary Jorgensen, Stewart
Sanderson, and Jeffrey Ott for assistance in data collection
and maintenance of the hybrid zone gardens.
Mountain
0
Basin
Hybrid
References _____________________
Mountain
Garden
70
Vegetative height (cm)
B
60
50
Soil
40
Basin
Hybrid
Mountain
30
Basin
Hybrid
Mountain
Garden
Height including inforescence (cm)
80
C
70
60
50
40
30
Soil
20
Basin
Hybrid
10
Mountain
0
N=
17
21
24
23
Basin
23
23
Hybrid
22
24
20
Mountain
Garden
70
Average crown diameter
D
60
50
Soil
40
Basin
Hybrid
Mountain
30
N=
19
21
Basin
25
23
23
23
Hybrid
22
24
20
Mountain
Garden
Figure 1—Mean (A) average inflorescence
length of plants, (B) vegetative height, (C)
vegetative height including inflorescence,
and (D) average crown diameter in each of
the three soils in each of the three gardens.
USDA Forest Service Proceedings RMRS-P-31. 2004
Arnold, M. L. 1997. Natural hybridization and evolution. New York:
Oxford University Press.
Barton, N. H. 1979a. Gene flow past a cline. Heredity. 44: 277–278.
Barton, N. H. 1979b. The dynamics of hybrid zones. Heredity. 43:
341–359.
Barton, N. H.; Hewitt, G. M. 1985. Analysis of hybrid zones. Annual
Review of Ecology and Systematics. 16: 113–148.
Barton, N. H.; Hewitt, G. M. 1989. Adaptation, speciation and
hybrid zones. Nature. 341: 497–503.
Beetle, A. A. 1960. A study of sagebrush, the section Tridentatae of
Artemisia. Bull. 368. Laramie: University of Wyoming, Agriculture Experiment Station. 83 p.
Beetle, A. A.; Young, A. 1965. A third subspecies in the Artemisia
tridentata complex. Rhodora. 67: 405–406.
Britch, S. C.; Cain, M. L.; Howard, D. J. 2001. Spatio-temporal
dynamics of the Allonemobius fasciatus-A. socius mosaic hybrid
zone: a 14-year perspective. Molecular Ecology. 10: 627–638.
Dobzhansky, T. 1940. Speciation as a stage in evolutionary divergence. American Naturalist. 74: 312–321.
Eckenwalder, J. E. 1984. Natural intersectional hybridization between North American species of Populus (Salicaceae) in sections
Aigeiros and Tacamhaca. III. Paleobotany and evolution. Canadian Journal of Botany. 62: 336–342.
Emms S. K.; Arnold, M. L. 1997. The effect of habitat on parental
and hybrid fitness: reciprocal transplant experiments with Louisiana irises. Evolution. 51: 1112–1119.
Freeman, D. C.; Miglia, K. J.; McArthur, E. D.; Graham, J. H.;
Wang, H. 1999a. Narrow hybrid zone between two subspecies of
big sagebrush (Artemisia tridentata: Asteraceae): X. Performance
in reciprocal transplant gardens. In: McArthur, E. Durant; Ostler,
W. Kent, Wambolt, L. Carl, comps. Proceedings: shrubland ecotones; 1998 August 12–14; Ephraim, UT. Proc. RMRS-P-11.
Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station: 15–24.
Freeman, D. C.; Wang, H.; Sanderson, S. C.; McArthur, E. D. 1999b.
Characterization of a narrow hybrid zone between two subspecies
of big sagebrush (Artemisia tridentata: Asteraceae). VII. Community and demographic analyses. Evolutionary Ecology Research.
1: 487–502.
Graham, J. H.; Freeman, D. C.; McArthur, E. D. 1995. Narrow
hybrid zone between two subspecies of big sagebrush (Artemisia
tridentata: Asteraceae). II. Selection and dispersal. American
Journal of Botany. 82: 709–716.
Graham, J. H.; McArthur, E. D.; Freeman, D. C. 2001a. Narrow
hybrid zone between two subspecies of big sagebrush (Artemisia
tridentata: Asteraceae). XI. Plant-insect interactions in reciprocal transplant gardens. In: McArthur, E. Durant; Fairbanks,
Daniel J., comps. Shrubland ecosystem genetics and biodiversity: proceedings; 2000 June 13–15; Provo, UT. Proc. RMRS-P21. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station: 118–126.
Graham, J. H.; McArthur, E. D.; Freeman, D. C. 2001b. Narrow
hybrid zone between two subspecies of big sagebrush (Artemisia
tridentata: Asteraceae). XII. Galls on sagebrush in a reciprocal
transplant garden. Oecologia. 126: 239–246.
Grant, V. 1963. The origins of adaptations. New York: Columbia
University Press.
Harrison, R. G. 1990. Hybrid zones: windows on evolutionary
process. Oxford Surveys in Evolutionary Biology. 7: 69–128.
35
Miglia, Freeman, McArthur, and Smith
Harrison, R. G.; Rand, D. M. 1989. Mosaic hybrid zones and the
nature of species boundaries. In: Otte, D.; Endler, J., eds. Speciation and its consequences. Sunderland, MA: Sinauer Associates:
111–113.
Hewitt, G. M. 1988. Hybrid zones—natural laboratories for evolutionary studies. Trends in Ecology and Evolution. 3: 158–167.
Howard, D. J.; Waring, G. L.; Tibbets, C. A.; Gregory, P. G. 1993.
Survival of hybrids in a mosaic hybrid zone. Evolution. 47: 789–800.
Linhart, Y. A.; Grant, M. C. 1997. Evolutionary significance of local
genetic differentiation in plants. Annual Review of Ecology and
Systematics. 27: 237–277.
Mayr, E. 1942. Systematics and the origin of species. New York:
Columbia University Press.
McArthur, E. D. 1994. Ecology, distribution, and values of sagebrush within the Intermountain region. In: Monsen, S. B.; Kitchen,
S. G., comps. Proceedings—ecology and management of annual
rangelands; 1992 May 18–21; Boise, ID. Gen. Tech. Rep. INTGTR-313. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station: 347–351.
McArthur, E. D.; Blauer, A. C.; Plummer, A. P.; Stevens, R. 1979.
Characteristics and hybridization of important Intermountain
shrubs. III. Sunflower family. Res. Pap. INT-220. Ogden, UT:
U.S. Department of Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station. 82 p.
McArthur, E. D.; Plummer, A. P. 1978. Biogeography and management of native western shrubs: a case study, section Tridentatae
of Artemisia. Great Basin Naturalist Memoirs. 2: 229–243.
McArthur, E. D.; Welch, B. L. 1982. Growth rate differences among
big sagebrush (Artemisia tridentata) accessions and subspecies.
Journal of Range Management. 35: 396–401.
McArthur, E. D.; Welch, B. L.; Sanderson, S. C. 1988. Natural and
artificial hybridization between big sagebrush (Artemisia
tridentata) subspecies. Journal of Heredity. 79: 268–276.
Moore, W. S. 1977. An evaluation of narrow hybrid zones in vertebrates. Quarterly Review of Biology. 52: 263–277.
Rand, D. M.; Harrison, R. G. 1989. Ecological genetics of a mosaic
hybrid zone: mitochondrial, nuclear, and reproductive differentiation of crickets by soil type. Evolution. 43: 432–449.
36
Importance of Genotype, Soil Type, and Location on the Performance ...
Rieseberg, L. H.; Baird, S. J. E.; Desrochers, A. M. 1998. Patterns of
mating in wild sunflower hybrid zones. Evolution. 52: 713–726.
Ross, C. L.; Harrison, R. G. 2002. A fine-scale spatial analysis of the
mosaic hybrid zone between Gryllus firmus and Gryllus
pennsylvanicus. Evolution. 56: 2296–2312.
Stebbins, G. L. 1950. Variation and evolution in plants. New York:
Columbia University Press.
SPSS Inc. 2001. Version 10.0 statistical software. Chicago, IL.
Trimble, S. 1989. The sagebrush ocean, a natural history of the
Great Basin. Reno: University of Nevada Press. 248 p.
Wambolt, C. L. 1996. Mule deer and elk foraging preference for 4
sagebrush taxa. Journal of Range Management. 49: 499–503.
Wang, H. 1996. Ecological analysis of a stable hybrid zone between
two big sagebrush subspecies (Artemisia tridentata ssp.
tridentata and A. t. ssp. vaseyana). Detroit, MI: Wayne State
University. 132 p. Dissertation.
Wang, H.; Byrd, D. W.; Howard, D. L.; McArthur, E. D.; Graham, J. H.;
Freeman, D. C. 1998. Narrow hybrid zone between two subspecies of big sagebrush (Artemisia tridentata: Asteraceae). V. Soil
properties. International Journal of Plant Sciences. 159: 139–
147.
Wang, H.; McArthur, E. D.; Freeman, D. C. 1999. Characterization
of a narrow hybrid zone between two subspecies of big sagebrush
(Artemisia tridentata: Asteraceae). IX. Elemental uptake and
niche separation. American Journal of Botany. 86: 1099–1107.
Wang, H.; McArthur, E. D.; Sanderson, S. C.; Graham, J. H.;
Freeman; D. C. 1997. Narrow hybrid zone between two subspecies of big sagebrush (Artemisia tridentata: Asteraceae). IV.
Reciprocal transplant experiments. Evolution. 51: 95–102.
Welch, B. L.; McArthur, E. D. 1986. Wintering mule deer preference
for 21 accessions of big sagebrush. Great Basin Naturalist. 46:
113–115.
Welch, B. L.; Jacobson, T. L. C. 1988. Root growth of Artemisia
tridentata. Journal of Range Management. 41: 332–334.
Whittam, T. G. 1989. Plant hybrid zones as sinks for pests. Science.
244: 1490–1493.
USDA Forest Service Proceedings RMRS-P-31. 2004
Revegetation of Saline Playa
Margins
Robert R. Blank
James A. Young
Abstract: New shrub recruitment in saline playa margins is
limited by extremely high osmotic potentials of the seedbed. In the
Eagle Valley playa near Fernley, NV, recruitment is rare and occurs
mostly in recently deposited eolian and flood-deposited sediments
of low osmotic potential. In most instances, however, sediment is of
insufficient thickness to support long-term growth. In 1990, as part
of a plant/soil relationship study in the eastern end of Eagle Valley
playa, soil pits were excavated by backhoe in an environment consisting of mounds occupied by Sarcobatus vermiculatus, Atriplex
lentiformis ssp. torreyi, and Allenrolfea occidentalis amid unvegetated interspaces. Soil pits were refilled, but depressions about 2 by
0.5 m in area to a depth of between 20 and 60 cm remained. Within
5 years, a thick veneer of eolian dust had accumulated in all the pits
and supported robust recruitment of S. vermiculatus, A. torreyi, and
A. occidentalis. By the year 2002, some shrubs were over 0.5 m in
stature. Excavating small depressions in saline playa environments
appears to be an effective revegetation technology provided the area
has a source of low osmotic potential eolian material.
Introduction ___________________
Creation of small depressions in the soil surface is a useful
seedbed preparation technique to facilitate seedling establishment in dry environments. The technology, alternately
named land imprinting, gouging, and pitting, has been
shown to enhance water availability to plants in the critical
early establishment phase (Munshower 1994; Whisenant
1999). Our purpose here is to report on the recruitment of
salt desert shrubs in depressions left from research activities on an extremely saline playa margin environment. We
hypothesize that the reason these depressions facilitated
shrub recruitment is a combination of enhanced water
availability and capture of seeds and eolian dust, which
serves as a low osmotic potential media for seed germination
and early establishment.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Robert R. Blank is Soil Scientist and James A. Young is Range Scientist,
USDA Agricultural Research Service, Exotic and Invasive Weed Research
Unit, 920 Valley Road, Reno, NV 89512, U.S.A., e-mail: blank@unr.nevada.edu
USDA Forest Service Proceedings RMRS-P-31. 2004
Methods _______________________
The study area is the Eagle Valley playa (39˚44' N, 119˚2' W)
just southeast of Fernley, NV. The western boundary was
the terminus of the Truckee River during pluvial periods
and consists of coarse-textured deltaic and reworked eolian
sands. Elevation is 1,234 m. The site is a gradient from
barren, flat, fine-textured, salt-encrusted sediments to a
higher coarser textured and less saline complex of reworked
beach material, eolian sands, and alluvial colluvial material. Based on monitor wells, the water table is less than 3 m
in most years. Vegetation occurs on mounds and is dominated by Allenrolfea occidentalis ([S. Watson] Kuntze),
Atriplex lentiformis ssp. torreyi ([S. Watson] H.M. Hall &
Clements), Sarcobatus vermiculatus ([Hook.] Torrey), and
by the grass Distichlis spicata ([L.] Greene). In the less
saline and coarse-textured beach and colluvial deposits,
vegetation is dominated by Atriplex confertifolia ([Torrey
and Frémont] S. Watson) and Sarcobatus baileyi ([Cov.]
Jepson). In 1990, we described a sequence of seven soils
along a transect encompassing the width of the mounded
area from the barren playa surface southeast to the less
saline upland interface (transect distance about 1.2 km). A
backhoe was used to excavate to a depth of approximately 3 m.
After description of soils, pits were refilled with stockpiled
soil, but settling created depressions averaging over 0.5 m in
depth and 2 by 0.5 m in area. The site was ignored for several
years, and when we returned in 1995, we were surprised at
the robust recruitment of salt desert shrubs from the depressions. Photos presented here were taken in May of 2002.
Results and Discussion __________
The playa margin community of which this study focuses
is an inhospitable environment for plant growth with extreme aridity and high salt content (fig. 1; Blank and others
1998). The surface soil seedbed is far too saline to allow seed
germination; total soil water potentials measured monthly
throughout a 2-year period average between –35 and –111
MPa for mounds microsite and between –25 and –86 MPa for
interspace site (Blank and others 1994). Recruitment does
occur along flood washes and impediments where eolian
dust accumulates. Our data suggest that these microsites
accumulate sediment of low enough osmotic potential to
support germination (table 1). However, the low osmotic
strength substrate is generally too thin to support plant
37
Blank and Young
Revegetation of Saline Playa Margins
Figure 1—General view of Eagle Valley playa looking
northeast from the study area. In the middle background
is barren, nearly level, lacustrine sediments. Surrounding
this central core is a complex mound-intermound
community extending outward and merging into alluvial
fans. The mounds are occupied by halophytic shrubs.
Distichlis spicata often occupies intermound areas.
Extensive measurements indicate that the total soil
water potential (mostly osmotic) is far too negative to
allow seed germination. New recruitment requires
microsites where low osmotic potential substrates can
accumulate.
growth until the roots can reach the water table and sustain
growth (Trent and others 1997).
Our soil excavations have allowed eolian dust to accumulate along with seeds and thereby facilitate plant recruitment (figs. 1 through 5). Other factors that may come into
play include (1) micro-meteorological benefits from increased
shading and (2) a greater chance of roots reaching the water
table. We suggest that creating small depressions in similar
landscapes may be a cost effective way to recruit new plants
into similar saline environments provided there is a source
of low osmotic potential wind-driven substrate and a source
of seeds.
Figure 2—Partially filled soil description pit in a very
saline portion of the study area. There is almost no
recruitment of new plants in this community. Most of the
established shrubs are nearly dead in this area, which
is difficult to judge in this black and white photograph.
Even in this inhospitable environment, recruitment of
Sarcobatus vermiculatus has occurred, fostered by the
accumulation of eolian sediments in the soil pit.
Figure 3—Robust plant of Atriplex lentiformis ssp.
torreyi that recruited into one of our soil description
pits. This is the largest shrub in any of our soil pits and
attests to how conditions in the soil pits facilitate rapid
plant growth.
Table 1—Electrical conductivity and theoretical osmotic potentials for
recent eolian sediment and underlying material.
Sample
Recent eolian
Underlying material
a
Electrical
conductivity
(dS m–1)
1.93 sd = 1.10
29.3 sd = 11.7
Based on 298 ∞K assuming osmoticum is NaCl.
38
Theoretical
osmotic potential
(MPa)a
0.046
.81
Figure 4—A soil description pit now nearly completely
filled in by eolian sediments. Notice the spoil pile in the
background. In this soil pit there was recruitment of
Sarcobatus vermiculatus, Allenrolfea occidentalis,
Atriplex lentiformis ssp. torreyi, and A. confertifolia.
USDA Forest Service Proceedings RMRS-P-31. 2004
Revegetation of Saline Playa Margins
Blank and Young
Figure 5—Soil description pit showing recruitment of
Sarcobatus vermiculatus and Atriplex confertifolia.
The more general top view shows very minor
recruitment in the surrounding area. Photo is taken
facing south and shows how eolian sediments
accumulate mainly on the west side of pits. Bottom
closeup photo shows that the depressions collect
seeds.
References _____________________
Blank, R. R.; Young, J. A. ; Martens, E.; Palmquist, D. E. 1994.
Influence of temperature and osmotic potential on germination
of Allenrolfea occidentalis. Journal of Arid Environments. 26:
339–347.
Blank, R. R.; Young, J. A.; Trent, J. D.; Palmquist, D. E. 1998.
Natural history of a saline mound ecosystem. Great Basin Naturalist. 58: 217–230.
USDA Forest Service Proceedings RMRS-P-31. 2004
Munshower, F. F. 1994. Practical handbook of disturbed land
revegetation. Boca Raton, FL: Lewis Publishers, Inc. 265 p.
Trent, J. D.; Blank, R. R.; Young, J. A. 1997. Ecophysiology of the
temperate desert halophytes: Allenrolfea occidentalis and
Sarcobatus vermiculatus. Great Basin Naturalist. 57: 57–65.
Whisenant S. G. 1999. Repairing damaged wildlands: a processoriented, landscape-scale approach. Cambridge, UK: Cambridge
University Press. 312 p.
39
Influence of Sagebrush and
Grass Seeding Rates on
Sagebrush Density and Plant
Size
Laurel E. Vicklund
Gerald E. Schuman
Ann L. Hild
Abstract: Wyoming big sagebrush (Artemisia tridentata ssp.
wyomingensis) establishment on mined lands in Wyoming is a
critical element of reclamation success. Because of its importance to
wildlife, shrub density is used to assess reclamation success for
wildlife habitat in Wyoming. Research was initiated at the Belle Ayr
Mine near Gillette in 1999 to evaluate the effects of Wyoming big
sagebrush seeding rates and grass competition on sagebrush establishment. Sagebrush seedling densities demonstrated consistent
increases with increased sagebrush seeding rates. The 4 kg PLS per
ha sagebrush seeding rate resulted in significantly greater density
than either the 2 or 1 kg PLS per ha seeding rate in 1999 to 2001.
Grass competition (grass seeding rates of 0, 2, 4, 6, 8, 10, and 14 kg
PLS per ha of a mixture of C3 native species) did not significantly
affect sagebrush seedling density. However, 1999 and 2000 precipitation was above or near normal, resulting in adequate moisture for
seedling emergence and growth of all seeded species. Sagebrush
seedling density declined at the higher grass seeding rates in 2001,
but seedling density was not significantly affected by grass seeding
rates (P = 0.12). Precipitation in 2001 was well below normal, and
grass competition significantly affected sagebrush seedling canopy
volume. Sagebrush seedling canopies were significantly smaller at
grass seeding rates greater than or equal to 4 kg PLS per ha. Continued evaluation of these treatments on sagebrush seedling performance will enable development of a seeding strategy that enhances
establishment of Wyoming big sagebrush on reclaimed mine lands.
Introduction ____________________
Mined land reclamationists in the Western United States
must meet many requirements to achieve successful reclamation. In August 1996, specific shrub density require-
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Laurel E. Vicklund is Senior Environmental Engineer, RAG Coal West,
Inc., Belle Ayr Mine, Gillette, WY 82717, U.S.A. Gerald E. Schuman is Soil
Scientist, High Plains Grasslands Research Station, USDA-ARS, Cheyenne,
WY 82009, U.S.A. FAX: (307) 637-6124; e-mail: gschuman@lamar.
colostate.edu. Ann L. Hild is Associate Professor, Department of Renewable
Resources, University of Wyoming, Laramie, WY 82071, U.S.A.
40
ments were imposed on mine reclamation in Wyoming
(Wyoming DEQ 1996). This standard required that the
dominant postmining shrub on lands identified as wildlife
2
habitat be reestablished at a density of 1 shrub per m on 20
percent of the disturbed landscape. Herbaceous species can
create significant competition for shrub seedlings (Eissentat
and Caldwell 1988; Schuman and others 1998; Williams and
others 2002). However, reclamation guidelines and regulations require topsoil stabilization immediately after the
topsoil is replaced using a permanent vegetation cover that
is at least as productive as premine conditions. Establishment of a perennial vegetative cover that protects the topsoil
resource from erosion can prevent the establishment of
shrubs and other difficult-to-establish native plants. However, establishment of both herbaceous and shrub species is
necessary to provide the desired plant community diversity
and appropriate species for wildlife habitat. This study was
conducted to further evaluate the establishment of Wyoming big sagebrush (Artemisia tridentata ssp. wyomingensis)
when sown at three rates with a mixture of cool-season native
grasses sown at seven seeding rates (competition levels).
Methods _______________________
The study site was located at the RAG Coal West, Inc.,
Belle Ayr Mine, 29 km southeast of Gillette, WY. Climate at
the site is continental, elevation is 1,460 m, mean air
temperature is 6.7 ∞C and the average annual precipitation
is 376 mm (Belle Ayr Coal Mine 2001). The premine vegetation of the area is a northern mixed-grass prairie primarily
comprised of cool- and warm-season grasses and Wyoming
big sagebrush. Soils are derived from Tertiary and Upper
Cretaceous shale, limestone, and sandstone. The topsoil was
a sandy clay loam, with a pH of 7.6, EC of 2.0 dS per m, total
nitrogen concentration of 700 mg per kg, and a soil organic
carbon content of 1.0 percent.
Topsoil replacement (56 cm) was completed in December
1997 to January 1998, and the area was seeded to barley
(Hordeum valgare var. “Steptoe”) in April 1998 to establish
a stubble mulch. In December 1998, seven grass seeding rate
treatments (0, 2, 4, 6, 8, 10, and 14 kg PLS [pure live seed]
per ha) were randomly drilled into 6.5 by 27 m main plots
within each of four replicate blocks. The grass mixture
(approximately equal seed numbers of each species) was
USDA Forest Service Proceedings RMRS-P-31. 2004
Influence of Sagebrush and Grass Seeding Rates on Sagebrush Density and Plant Size
Results and Discussion __________
Sagebrush seedling density data (1999 and 2000) and
aboveground plant biomass data (2000) were reported earlier by Williams and others (2002) but will be included in this
paper to demonstrate longer term trends and responses over
a wider range of climatic conditions. Precipitation in 2001
was considerably less than in 1999 and 2000, which were
near-normal or above-normal precipitation years (fig. 1).
Grass Seeding Rate Responses ___
Biomass of planted grasses in 2000 and 2001 showed
similar trends (fig. 2). In 2000, aboveground biomass of
planted grasses was significantly lower for grass seeding
rates of 0 and 2 kg PLS per ha and did not differ among grass
USDA Forest Service Proceedings RMRS-P-31. 2004
100
Monthly total
Precipitation data (mm)
25 Year avg.
80
60
40
20
0
J F M A M J J A S O N D
J F M A M J J A S O N D
1999
J F M A M J J A S O N D
2000
2001
Figure 1—Monthly precipitation for 1999 to 2001
and long-term average annual precipitation, Belle
Ayr Mine, RAG Coal West, Inc., Gillette, WY (Belle
Ayr Coal Mine 2001).
1200
2000
Grass biomass (kg / ha)
composed of three cool-season native perennials: western
wheat grass (Pascopyrum smithii [Rydb.] A. Love), thickspike
wheatgrass (Elymus lanceolatus [Scribner & J.G. Smith]
Gould), and slender wheatgrass (Elymus trachycaulus [Link]
Gould ex Shinners). Each grass seeding treatment plot was
subdivided into three, 6- by 9-m plots, which were randomly
seeded to one of three Wyoming big sagebrush seeding rates
(1, 2, or 4 kg PLS per ha) in March 1999.
2
Six 1-m quadrats were permanently marked in each
sagebrush seeding rate subplot in early 1999 before seedling
emergence. Sagebrush seedlings were counted within each
of these quadrats on June 30, August 3, August 31, and
October 25, 1999; June 5 and September 18, 2000; and June
20 and September 27, 2001. In 1999, a high density of
colonizing forb species (Kochia scoparia, Melilotus officinalis,
Salsola kali) were present on the site. To mimic reclamation
practices and to aid in the counting of sagebrush seedlings
the plots were mowed at about 15 to 18 cm in height and the
material removed from the plots. In July 2000 and 2001
herbaceous plant biomass at peak standing crop was deter2
mined by clipping four 0.18-m quadrats within each of the
84 sagebrush by grass seeding rate subplots. Aboveground
biomass was separated into planted grasses, exotic grasses
(Bromus japonicus, B. tectorum, E. junceus), and forbs.
Volunteer barley from the stubble mulch was included with
the exotic grasses but comprised a very small portion of the
total biomass.
In 2001, sagebrush seedling volume was estimated on all
of the seedlings counted within the permanent quadrats.
This parameter was included because of the visual perception of differences in sagebrush seedling size among grass
seeding rates. Sagebrush seedling volume was determined
by using a caliper to measure the plant canopy diameter at
its widest point, crown diameter perpendicular to the first
measurement, and the plant height. Plant volume was
calculated with the assumption that plant shape most closely
resembled an ellipse cone.
Analysis of variance was conducted on plant biomass,
sagebrush seedling density, and sagebrush seedling volume
using a split-split plot randomized block design (SAS Institute 1999). Mean separation was accomplished using Fisher’s
protected least significant difference procedures. All statistical analyses were evaluated at P < 0.05.
Vickland, Schuman, and Hild
2001
1000
a
800
0
a
a
a
400
200
a
a
600
a
a
a
a
a
b
c b
0
2
4
6
8
10
14
Grass seeding rate (kg PLS/ha)
Figure 2—Aboveground grass biomass for 2000
and 2001 at seven grass seeding rates. Belle Ayr
Mine, RAG Coal West, Inc., Gillete, WY (means
within a year across grass seeding rates with the
same lower case letter are not significantly different,
P < 9,95).
seeding rates of greater than or equal to 4 kg PLS per ha. In
2001, planted grass aboveground biomass did not differ
among grass seeding rates greater than or equal to 2 kg PLS
per ha. These data indicate that grass seeding rates of 2 to
14 kg PLS per ha can produce similar aboveground production in the second and third year after establishment. Therefore, grass seeding rates could be lowered considerably from
rates typically used by mine reclamation specialists and still
achieve the mandated production standards (equal to or
greater than premine). Reducing grass seeding rates could
aid natural recruitment of native grass, forb, and shrub
species that are less competitive (Fortier 2000; Stevenson
and others 1995). Because the grass species planted in this
41
Influence of Sagebrush and Grass Seeding Rates on Sagebrush Density and Plant Size
Sagebrush Seeding Rate Responses
Sagebrush seedling densities differed among the three
sagebrush seeding rates within each sampling date. Sagebrush seedling density for all three seeding rates was greatest on the June 2000 count and generally declined throughout the growing season due to seedling mortality (fig. 5). In
both years (2000 and 2001) sagebrush seedling density was
greater with increased sagebrush seeding rate. Highest
densities occurred at the 4 kg PLS per ha seeding rate. Even
the 1 kg PLS per ha seeding rate resulted in sagebrush
2
seedling densities greater than the 1 seedling per m shrub
standard. However, considering the long-term survival for
big sagebrush reported by Kiger and others (1987) and
Schuman and Belden (2002), seedling densities observed at
the 1 kg PLS per ha seeding rate in our study would not meet
the shrub standard after years 9 and 10, when the standard
is assessed. Kiger and others (1987) reported survival rates
160
Sagebrush volume (cm2)
study were a mixture of bunchgrass and rhizomatous species, soil protection was adequate to ensure soil stability.
Sagebrush seedling density after three growing seasons
was not significantly different among grass seeding rates
(fig. 3), although, sagebrush seedling density within the 14
kg PLS per ha grass seeding rate was less than half that
observed at the lower grass seeding rates. Grass seeding
rate did not affect sagebrush seedling density even though
precipitation (221 mm) in 2001 was below the long-term
average (336 mm) for the site. Therefore, it appears that
sagebrush seedling density did not respond to grass competition in the first 3 years of the study.
Aboveground sagebrush seedling size (canopy volume)
differed among the grass seeding rates (fig. 4). Canopy
volume of sagebrush seedlings was significantly smaller at
grass seeding rates of greater than or equal to 4 kg PLS per
ha than at the 0 and 2 kg PLS per ha grass seeding rate. No
differences in sagebrush seedling volume were observed for
grass seeding rates of 4 to 14 kg PLS per ha.
140
120
100
Sagebrush density (seedings / m2)
June '01
Oct '01
5
4
bc
60
40
cd
cd
cd
d
10
14
0
0
4
2
6
8
Grass seeding rate (kg PLS/ha)
Figure 4—Influence of grass seeding rate on Wyoming
big sagebrush canopy volume in 2001, Belle Ayr
Mine, RAG Coal West, Inc., Gillette, WY (means with
the same lower case letters are not significantly
different, P > 0.05).
12
2 kg PLS / ha
1 kg PLS / ha
11
4 kg PLS / ha
10
9
a
a
8
7
a
6
a
b
b
5
b
4
b
3
2
c
c
c
c
June '00
Sept '00
June '01
Oct. '01
1
Figure 5—Wyoming big sagebrush seedling density
at three sagebrush seeding rates averaged across
grass seeding rates, 2000 to 2001, Bellel Ayr Mine,
RAG Coal West, Inc., Gillette, WY (means within a
sampling date with the same lower case letter are not
significantly different, P > 0.05).
3
2
1
0
0
2
4
6
8
10
Grass seeding rate (kg PLS/ha)
Figure 3—Influence of seven grass seeding rates on
Wyoming big sagebrush seedling density in 2001,
Belle Ayr Mine, RAG Coal West, Inc., Gillette, WY
(sagebrush seedling density did not differ among
grass seeding rates, P > 0.05).
42
b
80
0
6
a
20
Sagebrush density (seedings / m2)
Vickland, Schuman, and Hild
14
of 28 to 32 percent after 11 years on a volunteer stand at a
coal mine in northwestern Colorado (believed to be mountain big sagebrush, Artemisia tridentata ssp. vaseyana).
However, Schuman and Belden (2002) reported a 59 percent
average survival rate across several cultural treatments
after 8 years in the Powder River Basin of Wyoming in a
Wyoming big sagebrush stand seeded at 2.2 kg PLS per ha.
Summary and Conclusions _______
Based on the observed sagebrush seedling densities as
affected by grass seeding rate it would be difficult to make
USDA Forest Service Proceedings RMRS-P-31. 2004
Influence of Sagebrush and Grass Seeding Rates on Sagebrush Density and Plant Size
foolproof recommendations for a grass seeding rate that
would provide a sagebrush density that meets the shrub
standard after 10 years. However, if one assesses the effect
of the grass seeding rate on sagebrush seedling density and
canopy growth in the context of the sagebrush seeding rates
and planted grass aboveground biomass, we suggest that
sagebrush seeding rate of 2 kg PLS per ha and a grass
seeding rate of 4 kg PLS per ha will likely meet management
goals on comparable sites. This recommendation is based on
the lack of differences in planted grass aboveground biomass
in the 6 to 14 kg PLS per ha grass seeding rates, while grass
seeding rates greater than or equal to 4 kg PLS per ha
greatly reduced canopy size of the sagebrush seedlings.
Early results indicate a trend (nonsignificant) of fewer
sagebrush seedlings above the 6 kg PLS per ha grass seeding
rate in 2001. We know that as grass seeding rates increase,
natural recruitment and establishment of difficult-to-establish native species are significantly reduced, particularly
desired forbs and shrubs (Bergelson and Perry 1989;
Eissenstat and Caldwell 1988; Richardson and others 1986).
We believe this recommendation should result in adequate
cover, provide adequate forage production, reduce vegetative competition, and enable establishment of Wyoming big
sagebrush to meet the shrub density standard required for
successful mine reclamation.
Acknowledgments ______________
We would like to thank Mary (Fortier) Williams, Rich
Vincent, and Darrell Ueckert for their review of this paper.
We also acknowledge the assistance of Mary Williams, Matt
Mortenson, Cliff Bowen, Kristene Partlow, Jennifer (Boyle)
Muscha, Leah Burgess, Krissie Peterson, Larry Griffith,
Pam Freeman, Ernie Taylor, Doug Miyamoto, Kelli Sutphin,
and Wes Brown and other technical support staff for help in
collecting the field data. We acknowledge support of RAG
Coal West, Inc., Belle Ayr Mine, Gillette, WY; the USDA
ARS High Plains Grasslands Research Station, Cheyenne,
WY; and the Department of Renewable Resources, University of Wyoming. This research was supported in part by the
USDA Forest Service Proceedings RMRS-P-31. 2004
Vickland, Schuman, and Hild
Wyoming Abandoned Coal Mine Land Research Program,
administered by the Office of Research, University of Wyoming, and the Abandoned Mine Land Division, Wyoming
Department of Environmental Quality, Cheyenne.
References _____________________
Belle Ayr Coal Mine. 2001. Belle Ayr mine annual report. RAG Coal
West, Inc., Gillette, WY.
Bergelson, J.; Perry, R. 1989. Interspecific competition between
seeds: relative planting date and density affect seedling emergence. Ecology. 70: 1639–1644.
Eissenstat, D. M.; Caldwell, M. M. 1988. Competitive ability is
linked to rates of water extraction: a field study of two aridland
tusssock grasses. Oecologia. 75: 1–7.
Fortier, M. I. 2000. Wyoming big sagebrush establishment on mined
lands: seeding rates and competition. Laramie: University of
Wyoming. Thesis.
Kiger, J. A.; Berg, W. A.; Herron, J. T.; Phillips, C. M.; Atkinson, R. G.
1987. Shrub establishment in the mountain shrub zone. In:
Proceedings, 4th biennial symposium on surface mining and
reclamation of the Great Plains and 4th annual meeting of the
American Society for Surface Mining and Reclamation; 1987
March 16–20; Billings, MT. Montana Reclamation Research Unit
Report 8704. Bozeman: Montana State University: L-3-1 to L-3-6.
Richardson, B. Z.; Monsen, S. B.; Bowers, D. M. 1986. Interseeding
selected shrubs and herbs on mine disturbance in southeastern
Idaho. In: McArthur, E. D.; Welch, B. L., comps. Proceedings:
symposium on the biology of Artemisia and Chrysothamnus. Gen.
Tech. Rep. INT-200. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Research Station: 134–139.
SAS Institute. 1999. SAS/STAT user’s guide, release 8.0. Cary, NC.
Schuman, G. E.; Belden, S. E. 2002. Long-term survival of direct
seeded Wyoming big sagebrush seedlings on a reclaimed mine
site. Arid Land Research and Management. 16: 309–317.
Schuman, G. E.; Booth, D. T.; Cockrell, J. R. 1998. Cultural methods
for establishing Wyoming big sagebrush on mined lands. Journal of Range Management. 51: 223–230.
Stevenson, M. J.; Bullock, J. M.; Ward, L. K. 1995. Re-creating seminatural communities: effect of sowing rate on establishment of
calcareous grassland. Restoration Ecology. 3: 279–289.
Williams, M. I.; Schuman, G. E.; Hild, A. L.; Vicklund, L. E. 2002.
Wyoming big sagebrush density: effects of seeding rates and
grass competition. Restoration Ecology. 10: 385–391.
Wyoming Department of Environmental Quality, Land Quality
Division. 1996. Coal rules and regulations. Chapter 4, appendix A.
State of Wyoming, Cheyenne.
43
Soil Components and
Microsites
Eriogonum exilfiolium
45
Illustration on the reverse side is by Walt Fertig. In: Wyoming Natural Diversity Database. 2000. Available at: http://uwadmnweb.uwyo.edu/wyndd
(Accessed 12/18/03).
Soil Nitrogen Controls on Grass
Seedling Tiller Recruitment
Thomas A. Monaco
Jay B. Norton
Douglas A. Johnson
Thomas A. Jones
Jeanette M. Norton
Abstract: Establishment of perennial grass seedlings is critical for
repairing rangeland ecosystems in the Western United States.
Mineral N in the soil may be one of the controlling factors for
seedling establishment and may improve the competitive ability of
perennial grasses compared to the annual grass, cheatgrass (Bromus
tectorum L.). A greenhouse experiment was conducted to investigate the effects of immobilizing mineral soil N on tiller recruitment
in grass seedlings. The perennial grasses big squirreltail (Elymus
multisetus [J.G. Smith] M.E. Jones), “Goldar” bluebunch wheatgrass (Pseudoroegneria spicata [Pursh] A. Löve), and “CD II” crested
wheatgrass (Agropyron cristatum [L.] Gaertner x A. desertorum
[Fisch. ex Link] Schultes) were grown alone or with cheatgrass in
pots for 7 weeks. Plants were grown in soil amended with ground
barley straw to produce the following four treatments designed to
immobilize mineral N: (1) no straw added (control), (2) 0.25 mg straw
kg–1 soil, (3) 0.50 mg straw kg–1 soil, and (4) 1.00 mg straw kg–1 soil.
Number of tillers for perennial grasses grown alone was not significantly different than perennial grasses grown with cheatgrass.
Cheatgrass and crested wheatgrass plants grown alone produced
significantly fewer (36 to 39 percent) new tillers per plant in the 1.00
mg kg–1 treatment compared to the control. Squirreltail had significantly fewer tillers per plant in the highest straw treatment
compared to the control when grown with cheatgrass. Cheatgrass
produced more than twice as many tillers as the perennial grasses
both in the control (high mineral N) and in the high straw treatment
(low mineral N). Soil in pots containing cheatgrass had 94 percent
less mineral N than soil in pots containing bluebunch wheatgrass.
These results suggest that cheatgrass will likely be an effective
competitor with perennial grasses even if mineral soil N is drastically reduced by microbial immobilization.
Introduction ___________________
Establishing perennial grasses in rangelands degraded
by invasive annual grasses continues to be one of the
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Thomas A. Monaco, Douglas A. Johnson, and Thomas A. Jones, USDAARS Forage and Range Research Laboratory, Utah State University, Logan,
UT 84322-6300, U.S.A. Jay B. Norton and Jeanette M. Norton, Utah State
University, Plants, Soils, and Biometeorology Department, Logan, UT 843224820, U.S.A.
USDA Forest Service Proceedings RMRS-P-31. 2004
greatest challenges to repairing ecosystems in the Intermountain West. Many characteristics of the invasive annual
cheatgrass (Bromus tectorum L.) allow it to be a superior
competitor for belowground soil resources (Harris 1967). A
better understanding of how changes in mineral N availability alters species competitive interactions between cheatgrass
and perennial grasses (Grime and others 1987; Tilman
1984) may lead to new weed management strategies for
enhancing perennial grass establishment.
Annual plants such as cheatgrass generally have a higher
growth rate than longer lived perennial grasses (Arredondo
and others 1998). In addition, annual grasses have higher
NO3– uptake and greater N productivity (Garnier and
Vancaeyzeele 1994; Poorter and others 1990), which may
reduce the success of seeded perennial grasses on semiarid
rangelands (Hironaka 1961). High growth rate and mineral
N uptake allow cheatgrass seedlings to emerge earlier (for
example, Pyke 1990) and exhibit greater early spring root
growth and root proliferation in fertilized soil microsites
(Caldwell and others 1991; Eissenstat and Caldwell 1988)
than native perennial grasses. Recent evidence suggests
that low concentrations of mineral N hinder potential shoot
and root growth of both cheatgrass and native perennial
grasses (Monaco and others 2003). However, it is uncertain
whether reducing mineral N will improve the establishment
of perennial grasses under competition with cheatgrass.
We conducted a greenhouse experiment with potted plants
to evaluate the hypothesis that low mineral N would improve the competitive ability of perennial grasses when
grown with cheatgrass. Specifically, we anticipated that
treatments designed to reduce soil mineral N would hinder
juvenile tiller recruitment of cheatgrass relatively more
than that of perennial grasses. Number of tillers were
evaluated because tiller numbers directly affect numerous
demographic and ecological processes including plant size,
competitive ability, productivity, grazing resistance, and
population persistence (Briske 1991). In seedlings of perennial and annual grasses, tiller number provides an effective
measure of a plant’s potential growth and establishment.
Materials and Methods __________
The effects of low mineral N on tiller recruitment in
cheatgrass, squirreltail, bluebunch wheatgrass, and crested
wheatgrass were evaluated in a greenhouse experiment in
Logan, UT. A coarse-loamy soil characterized as a Xeric
47
Monaco, Norton, Johnson, Jones, and Norton
Soil Nitrogen Controls on Grass Seedling Tiller Recruitment
Torriorhent was excavated from Dugway Proving Grounds
(40∞ 14’ 23” N, 112∞ 50’ 47” W) in Tooele County, UT, to a
maximum depth of 60 cm. The top 2 cm of soil and litter were
discarded to remove existing seeds in the upper soil layer.
Total soil carbon (C) (13 g kg–1) and total soil N (1.0 g kg–1)
were determined by direct combustion. The soil was passed
through a 6-mm sieve to remove rocks and organic debris,
and then thoroughly mixed.
Four soil treatments were established by mixing 7 kg of
soil with 0, 0.25, 0.50, and 1.00 mg kg–1 of ground barley
straw in a 5-gallon bucket and then placing the individual
soils into 8-L plastic pots. The 0 mg kg–1 straw treatment
served as a control. The straw treatment was designed to
decrease overall mineral N (NH4+, NO2–, and NO3–) availability by promoting microbial immobilization of mineral N
with a high C:N ratio organic material. The C:N ratio of the
straw was 98 as measured by direct combustion. Preliminary experiments indicated nearly all (>95 percent ) extractable soil mineral N was immobilized by the 1.00 mg kg–1 level
within 5 days when soils were maintained at field capacity
(6.9 percent soil water content). Mineral N concentrations
(mg kg–1) were 0.75 ± 0.01 (n = 8) in the control and 0.07
(n = 8) in the 1.00-mg-kg–1 straw treatment. All pots were
watered and weighed to reach field capacity for 5 consecutive
days before seedlings were transplanted into pots.
Soil mineral N concentrations were analyzed according to
the methods described by Hart and others (1994). A 10-g soil
sample was taken, homogenized, and extracted with 2 M
KCl within 4 hours. Extracts were filtered through filter
paper preleached with 2 M KCl and frozen until analyzed.
Concentrations of NO2– + NO3– and NH4+ were analyzed
colorimetrically with a flow injection autoanalyzer using
standard procedures (Lachat 1989, 1990).
Seeds of each grass were germinated on blotter paper, and
four seedlings were transplanted into each pot. Grasses
were planted alone or in a 2:2 mixture with cheatgrass. Each
grass x mixture x treatment combination was replicated four
times. Pots were randomly arranged on greenhouse benches
and grown for 7 weeks under 50-percent ambient solar
radiation to help maintain air temperature at 70 ∞C during
the experiment in late spring and summer (May to June
2001). Plants were watered each day with the amount of
water required to reach field capacity. After 7 weeks, the
number of tillers produced was counted on all plants. Many
plants of each grass appeared to be N deficient in the straw
treatments. As a result, all pots were fertilized at 7 weeks
with 500 ml of NPK (20-20-20) fertilizer to revive active
vegetative growth. After an additional 6 weeks of growth,
soil mineral N was analyzed for the control pots of grasses
grown without cheatgrass competition (n = 4).
Tiller numbers were analyzed individually for each species with analysis of variance to determine the effects of
straw treatment and species mixture. A Tukey’s test was
performed on treatment means when a significant main
effect was observed. All statistical tests were performed with
a = 0.05.
Results and Discussion _________
No significant differences in tiller numbers were observed
between perennial grass plants grown alone and those
grown in competition with cheatgrass (table 1). In contrast,
the high straw treatment significantly reduced the number
of tillers compared to the control for squirreltail when grown
with cheatgrass and for crested wheatgrass when grown
alone. Immobilizing mineral N with straw also significantly
reduced tiller numbers of cheatgrass plants grown alone
(table 2). Of the four grasses we evaluated, bluebunch
wheatgrass was the only grass that did not show a significant decrease in tiller numbers under low mineral N. Our
results do not support our initial hypothesis that reducing
mineral N would improve the establishment of perennial
grasses with cheatgrass competition.
Two alternative hypotheses may account for our results
that tiller numbers of cheatgrass, squirreltail, and crested
wheatgrass were negatively impacted by low soil concentrations of mineral N, while tiller numbers for bluebunch
wheatgrass were not. One possible hypothesis is that high
growth rates in the three responsive grasses could not be
supported when mineral N was limiting (Berendse and
others 1992; van der Werf and others 1993a,b). Bluebunch
wheatgrass, which is known to have a lower growth rate
than the other three grasses (Arredondo and others 1998),
may have been the least responsive grass to the straw
treatments because it is likely functioning closer to its
“optimal growth and metabolic rate” when soil N is limiting
(for example, Chapin and others 1987).
A second hypothesis that may partially explain the tiller
number responses we observed is that fast-growing invasive
annual grasses and slow-growing perennial grasses may
allocate available soil N differently to shoots and roots. For
Table 1—Mean (± 1 SE, n = 4) number of tillers per plant for squirreltail, bluebunch wheatgrass, and crested
wheatgrass when grown alone and in equal mixture with cheatgrass. Means within a row followed by the
same letters are not significantly different.
Treatment
Species
48
Control
0.25
0.50
1.00
Squirreltail
Alone
with cheatgrass
- - - - - - - - - - - - - mg barley straw kg-1 soil - - - - - - - - - - - - - - 3.8 ± 0.4
4.4 ± 0.7
3.6 ± 0.6
2.8 ± 0.6
5.0 ± 1.0a
3.1 ± 0.3ab
2.8 ± 0.4ab
2.1 ± 0.4b
Bluebunch wheatgrass
Alone
with cheatgrass
3.8 ± 0.7
2.8 ± 0.6
2.3 ± 0.3
2.8 ± 0.3
2.3 ± 0.4
3.3 ± 0.8
2.0 ± 0.6
3.8 ± 1.0
Crested wheatgrass
Alone
with cheatgrass
6.3 ± 0.3a
6.6 ± 0.6
4.9 ± 0.1ab
3.9 ± 0.6
4.3 ± 0.3b
5.0 ± 0.5
4.0 ± 0.5b
4.9 ± 0.1
USDA Forest Service Proceedings RMRS-P-31. 2004
Soil Nitrogen Controls on Grass Seedling Tiller Recruitment
Monaco, Norton, Johnson, Jones, and Norton
Table 2—Mean (± 1 SE) number of tillers for cheatgrass plants when grown alone and in equal
mixture with squirreltail, bluebunch wheatgrass, and crested wheatgrass for 6 weeks.
Means within a row followed by the same letters are not significantly different.
Cheatgrass
n
Alone
With squirreltail
With bluebunch wheatgrass
With crested wheatgrass
12
4
4
4
----------11.2 ± 0.8a
14.5 ± 1.9
12.5 ± 1.7
12.1 ± 2.3
–1
(mg kg soil)
–
+
Soil N (NO3 + NH4 )
example, Monaco and others (2003) found that invasive
annual grasses primarily allocate N and biomass to shoots,
whereas allocation was primarily to roots in the native
perennial bluebunch wheatgrass. Thus, cheatgrass,
squirreltail, and crested wheatgrass may have responded to
reductions in soil N by altering aboveground structures
(tillers), while low N conditions may have had greater
impacts on root growth in bluebunch wheatgrass.
High growth rates and different shoot and root allocation
patterns may have collectively enabled cheatgrass,
squirreltail, and crested wheatgrass to better exploit the
limiting N supply in pots and perform relatively better
than or equal to bluebunch wheatgrass (fig. 1). The amount
of soil N remaining in pots occupied by these grasses
indicated that cheatgrass pots had the least amount of
mineral N followed by crested wheatgrass, squirreltail,
and bluebunch wheatgrass. Pots containing cheatgrass
had 94 percent less mineral N than bluebunch wheatgrass.
High consumption of mineral N by cheatgrass likely facilitated its production of more than twice as many tillers as the
perennial grasses, both in the control (high mineral N) and in
the high straw treatment (low mineral N). However, it remains unclear why tiller numbers for perennial grasses
grown alone were not significantly different than when
60
a
ab
40
ab
20
b
0
Bluebunch
WG
Squirreltail
Crested
WG
Cheatgrass
Species
Figure 1—Mean (± 1 SE, n = 4) mineral N (NO3– and
NH4+) of soils containing bluebunch wheatgrass
(WG), squirreltail, crested wheatgrass (WG), and
cheatgrass. Vertical bars labeled with different
lowercase letters are significantly (P < 0.05)
different.
USDA Forest Service Proceedings RMRS-P-31. 2004
Treatment
0.25
0.50
Control
1.00
-1
mg barley straw kg soil - - - - - - - - - - 9.2 ± 0.6ab
7.5 0.6b
6.8 ± 0.6b
8.3 ± 2.1
9.0 ± 1.8
5.9 ± 1.3
11.6 ± 2.2
7.3 ± 0.8
9.1 ± 1.0
8.4 ± 0.8
9.6 ± 1.1
7.0 ± 0.7
grown with cheatgrass, given the ability of cheatgrass to
modify mineral N availability. Additional insight into the
question of whether competitive ability of perennial grasses
increases relative to cheatgrass under low mineral N may
become apparent with more detailed experiments (Gibson
and others 1999). Our results suggest that cheatgrass will likely
be an effective competitor with perennial grasses even if mineral soil N is drastically reduced by microbial immobilization.
Acknowledgments _____________
Research was funded by USDA-ARS and USDA-CSREES
Grant No. 97-38300-4892. We thank Justin Williams, Brandon Gordon, Kevin Connors, and Jacqueline Adams for
assisting with plant maintenance and data collection.
References ____________________
Arredondo, J. T.; Jones, T. A.; Johnson, D. A. 1998. Seedling growth
of Intermountain perennial and weedy annual grasses. Journal
of Range Management. 51: 584–589.
Berendse, F.; Elberse, W. Th.; Geerts, R. H. M. E. 1992. Competition
and nitrogen loss from plants in grassland ecosystems. Ecology.
73: 46–53.
Briske, D. D. 1991. Developmental morphology and physiology of
grasses. In: Heitschmidt, R. K.; Stuth, J. W., eds. Grazing management: an ecological perspective. Portland, OR: Timber Press:
85–108.
Caldwell, M. M.; Manwaring, J. H.; Durham, S. L. 1991. The
microscale distribution of neighbouring plant roots in fertile soil
microsites. Functional Ecology. 5: 765–772.
Chapin, F. S., III; Bloom, A. J.; Field, C. B.; Waring, R. H. 1987.
Plant responses to multiple environmental factors. Bioscience.
37: 49–57.
Eissenstat, D. M.; Caldwell, M. M. 1988. Seasonal timing of root
growth in favorable microsites. Ecology. 69: 870–873.
Garnier E.; Vancaeyzeele, S. 1994. Carbon and nitrogen content of
congeneric annual and perennial grass species: relationships
with growth. Plant, Cell and Environment. 17: 399–407.
Gibson, D. J.; Connolly, J.; Hartnett, D. C.; Weidenhamer, J. D.
1999. Designs for greenhouse studies of interactions between
plants. Journal of Ecology. 87: 1–16.
Grime J. P.; Mackey, J. M. L.; Hillier, S. H.; Read, D. J. 1987.
Floristic diversity in a model system using experimental microcosms. Nature. 328: 420–422.
Harris, G. A. 1967. Some competitive relationships between Agropyron spicatum and Bromus tectorum. Ecological Monographs.
37: 89–111.
Hart, S. C.; Stark, J. M. E.; Davidson, A.; Firestone, M. K. 1994.
Nitrogen mineralization, immobilization and nitrification. In:
Weaver, R. W.; Angle, S.; Bottomley, P.; Bezdicek, D.; Smith, S.;
49
Monaco, Norton, Johnson, Jones, and Norton
Tabatabai, A.; Wollum, A., eds. Methods of soil analysis microbiological and biochemical properties. 3d ed. Madison, WI: Soil
Science Society of America: 985–1018.
Hironaka, M. 1961. The relative rate of root development of
cheatgrass and medusahead. Journal of Range Management.
14: 263–267.
Lachat. 1989. Operations manual for the QuikChem automated ion
analyzer. Quikchem 12-107-04-1-B (nitrate). Milwaukee, WI:
Lachat.
Lachat. 1990. Operations manual for the QuikChem automated
ion analyzer. Quikchem 12-107-06-2-A (ammonium). Milwaukee, WI: Lachat.
Monaco, T. A.; Johnson, D. A.; Norton, J. M.; Jones, T. A.; Connors,
K. J.; Norton, J. B.; Redinbaugh, M. B. 2003. Contrasting responses of Intermountain West grasses to soil nitrogen. Journal
of Range Management. 56: 282–290.
50
Soil Nitrogen Controls on Grass Seedling Tiller Recruitment
Poorter, H.; Remkes, C.; Lambers, H. 1990. Carbon and nitrogen
ecology of 24 wild species differing in relative growth rate. Plant
Physiology. 94: 621–627.
Pyke, D. A. 1990. Comparative demography of co-occurring introduced and native tussock grasses: persistence and potential
expansion. Oecologia. 82: 537–543.
Tilman, D. 1984. Plant dominance along an experimental nutrient
gradient. Ecology. 65: 1445–1453.
van der Werf, A.; van Nuenen, M.; Visser, A. J.; Lambers, H. 1993a.
Contribution of physiological and morphological plant traits to a
species’ competitive ability at high and low nitrogen supply.
Oecologia. 94: 434–440.
van der Werf, A.; Visser, A. J.; Schieving, F.; Lambers, H. 1993b.
Evidence for optimal partitioning of biomass and nitrogen at a
range of nitrogen availabilities for fast and slow-growing species.
Functional Ecology. 7: 63–74.
USDA Forest Service Proceedings RMRS-P-31. 2004
Enzyme Activity in Temperate
Desert Soils: Influence of
Microsite, Depth, and Grazing
Robert R. Blank
Abstract: The enzyme status of soil influences mineralization
kinetics, and thus, the supply of nutrients to plants. We quantified
urease, asparaginase, glutaminase, phosphatase, and arylsulfatase
activity in a sagebrush/grass ecosystem northeast of Reno, NV.
Enzyme activity was evaluated by depth (0 to 5 cm, 5 to 10 cm, 10
to 20 cm), microsite (sagebrush, crested wheatgrass, cheatgrass,
cryptogamic crust in shrub interspace, noncryptogamic crust in
shrub interspace), and treatment (grazed and ungrazed). For most
enzymes evaluated, there was a significant depth x microsite interaction. In general, enzyme activity declined with depth. Moreover,
the noncryptogamic interspace microsite often had the lowest
enzyme activity among the other microsites. Depending on soil
depth and microsite, the grazing treatment significantly reduced
urease, asparaginase, and alkaline phosphatase activity compared
to the ungrazed treatment. Enzyme activity is an important soil
attribute and may serve as a robust measure of soil health.
Introduction ___________________
Soil is a critical component of the Earth’s biosphere. From
food production, degradation of toxic compounds, and as a
medium for the geochemical recycling of many elements,
proper management of soil is crucial for the continued
prosperity of humans. Yet, given that most soils of the world
have only been intensively cultivated and grazed for a
relatively short period of time, there are concerns whether
our soil resources are sustainable (Hillel 1991). Scientists
have recently attempted to quantify soil health or quality as
an index of sustainability (Doran and others 1994). One
potentially important soil attribute that may be a good
proxy for soil health is enzyme activity (Dick 1994; Saviozzi
and others 2001) because it is an integration of life processes
occurring within the soil. The study was conducted to
ascertain how several common soil enzymes vary by soil
depth, soil microsite, and grazing in a sagebrush/grass plant
community.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Robert R. Blank is Soil Scientist, USDA Agricultural Research Service,
Exotic and Invasive Weed Research Unit, 920 Valley Road, Reno, NV 89512
U.S.A., e-mail: blank@unr.nevada.edu
USDA Forest Service Proceedings RMRS-P-31. 2004
Methods ______________________
Field
The study area is in the Bedell Flat area, 30 km north of
o
o
Reno, NV (119 50’ E 39 51’ N). The slope is less than 4 percent
with a northeastern aspect. Elevation is 1,548 m. The soil is
mapped as the Bedell series, a coarse-loamy, mixed, mesic,
Aridic Argixeroll developed in mixed colluvium dominantly
from granite. Present vegetation consists of crested wheatgrass (Agropyron desertorum [Fischer] Schultes), cheatgrass
(Bromus tectorum L.), big sagebrush (Artemisia tridentata
ssp. tridentata Nutt.), needle-and-thread (Hesperostipa
comata Trin. & Rupr.), Thurber’s needlegrass (Achnatherum
thurberianum [Piper] Barkworth), squirreltail (Elymus
elymoides [Raf.] Swezey), and green rabbitbrush
(Chrysothamnus viscidiflorus [Hook.] Nutt.). On July 19,
1995, soil (four replicates) was collected from grazed and
ungrazed treatment plots. The plots are on either side of an
exclosure constructed in 1958. Three depths were sampled (0 to
5 cm, 5 to 10 cm, and 10 to 20 cm). Microsites sampled on the
grazed plot included sagebrush, (ARTR), cheatgrass, (BRTE),
and noncryptogamic barren shrub interspaces (INTER).
Microsites sampled in the ungrazed plot included crested wheatgrass, (AGDE), cryptogamic crust covered shrub interspaces
(CRPTO), cheatgrass (BRTE), and sagebrush (ARTR).
Laboratory
The soils were returned to the laboratory where they were
air dried and sieved to remove material greater than 2 mm
in size. The soils were then stored in paper bags in the
refrigerator prior to the particular enzyme assays. Assays
were completed within 2 weeks. Enzyme activity procedures
are as outline in Tabatabai (1994). Three enzymes that
cleave amine groups (amidohyrolases) were evaluated: asparaginase, urease, and glutaminase. These assays are
based on the determination of ammonium released when
buffered soil (5 g) is incubated individually with known
amounts of the substrates, L-glutamine, L-asparagine, or
urea at 37 ∞C for 2 hours. After incubation, cleaved ammonium is extracted with 2 M KCl containing AgSO4 to stop
enzyme activity. Ammonium present in the original soil and
ammonium cleaved due to non-enzymatic processes are
subtracted out via running a blank. Ammonium is quantified using a flow injection membrane-diffusion method.
Quantification of arylsulfatase, and acid and alkaline
phosphatase activity is based on the cleaving of the sulfate or phosphate group attached to p-nitrophenyl. One g of
THAM-buffered soil is incubated with the appropriate
51
Blank
Enzyme Activity in Temperate Desert Soils: Influence of Microsite, Depth, and Grazing
substrate at 37 ∞C for 1 hour. The p-nitrophenyl remaining
is extracted with KCl and quantified colorimetrically.
Statistics
For each individual enzyme, data were analyzed using a
fixed effect, two-way analysis of variance with unequal
replication for depth and microsite. A separate two-way
analysis of variance was performed on depth and treatment
for microsites CRPTO/INTER, BRTE, and ARTR. If the
analysis showed a significant F-value, mean separation was
accomplished using Duncan’s new multiple range test. For
the prepared graphs, variance of data is shown by standard
error about the mean.
Results _______________________
There was a significant (P = 0.05) depth x microsite interaction for enzyme activities of urease, asparaginase,
arylsulfatase, and glutaminase (fig. 1). Urease and arylsulfatase activities were more variable than asparaginase or
glutaminase activity. Moreover, the INTER microsite generally has the lowest enzyme activities among the microsites
and the most inconsistent trend in enzyme activities with
depth. Urease activity displayed a declining trend with
depth for the CRPTO, ARTR, and BRTE microsites, but
activity for the INTER site significantly increased with
depth. Asparaginase was the only enzyme whose activity
consistently declined with depth. Arylsulfatase activity was
inconsistent with depth. The ARTR microsite had the highest enzyme activities, but only statistically higher in the
0- to 5-cm depth increment for urease and asparaginase.
Alkaline phosphatase had significant main effects for
depth and microsite (fig. 2). Enzyme activity declined significantly with depth. The CRPTO microsite had the most
activity followed by ARTR, AGDE, BRTE, and finally INTER with the least enzyme activity.
Asparaginase and urease activities were influenced by a
significant microsite x treatment interaction (fig. 3). In the
0- to 5-cm depth increment, the grazed plot had significantly
less enzyme activities of asparaginase and urease in the
INTER microsite compared to the ungrazed CRPTO
microsite. In addition, for the BRTE microsite, urease activity was significantly less in 10- to 20-cm depth increments of
the grazed plot compared to the corresponding ungrazed
plot. Finally, alkaline phosphatase activity was significantly less on the grazed INTER microsite than on the
ungrazed CRPTO sites.
Discussion ____________________
Plant microsite did significantly affect enzyme bioassays,
which has been reported in the literature (Neal 1973; Saviozzi
and others 2001). In general, enzyme activity was highest in
Figure 1—Soil enzyme activities that were influenced by a significant soil depth by
microsite interaction. Units are mmol of substrate cleaved g–1 hr–1.
52
USDA Forest Service Proceedings RMRS-P-31. 2004
Enzyme Activity in Temperate Desert Soils: Influence of Microsite, Depth, and Grazing
Figure 2—Phosphatase activity main effects of
microsite and soil depth. Units are mmol of
substrate cleaved g–1 hr–1.
Blank
some moderation of the intense desert sun and engender
greater microbial numbers, may explain these findings.
Grazing-induced reduction of enzyme activities has been
reported in the literature (Holt 1996). Given that enzyme
activity of soil is both a microbial (largely) and a plant
mediated process (Tate 1995), one would suspect that grazing has reduced those microbes and higher plants that
produce urease, asparaginase, and alkaline phosphatase.
This conclusion is of course complicated by the fact that, in
the shrub interspaces where the grazing effect is most
pronounced, the soil lacks well-expressed cryptogamic organisms. The large reduction in urease activity on grazed
BRTE microsites is perplexing; however, it seems reasonable to speculate that grazing of cheatgrass may have
reduced root elongation into the 10- to 20-cm depth increment. It seems plausible that the grazing reduction of the
amidohydrolases urease and asparaginase could potentially
reduce N mineralization kinetics (Holt 1997). Likewise grazing-induced reduction of phosphatase could reduce P availability. We are planning future experiments to clarify these
unknowns.
References ____________________
Figure 3—Soil enzyme activities that were
influenced by a significant main effect or
interaction with treatment. Units are mmol of
substrate cleaved g–1 dry soil hr–1.
the surface soil (0 to 5 cm). Burke and others (1989) and
Bolton and others (1993), whose experiments were also
conducted in a sagebrush ecosystem, found similar results.
For the cryptogamic crust and sagebrush interspace sites,
however, enzyme activity was least in the 0- to 5-cm depth
increment. The lack of higher plant life, which would afford
USDA Forest Service Proceedings RMRS-P-31. 2004
Bolton, H., Jr.; Smith, J. L.; Link, S. O. 1993. Soil microbial biomass
and activity of a disturbed and undisturbed shrub-steppe ecosystem. Soil Biology and Biochemistry. 25: 545–552.
Burke, I. C. 1989. Control of nitrogen mineralization in a sagebrush
steppe landscape. Ecology. 70: 1115–1126.
Dick, R. P. 1994. Soil enzyme activities as indicators of soil quality.
In: Defining soil quality for a sustainable environment. SSSA
Spec. Publ. 35. Madison, WI: Soil Science Society of America:
107–124
Doran, J. W.; and others, eds. 1994. Defining soil quality for a
sustainable environment. SSSA Spec. Publ. 35. Madison, WI: Soil
Science Society of America. 244 p.
Hillel, D. J. 1991. Out of the Earth, civilization and the life of the
soil. New York: The Free Press. 321 p.
Holt, J. A. 1997. Grazing pressure and soil carbon, microbial
biomass and enzyme activities in semi-arid Northeastern Australia. Applied Soil Ecology. 5: 143–149.
Neal, J. L., Jr. 1973. Influence of selected grasses and forbs on soil
phosphatase activity. Canadian Journal of Soil Science. 53:
119–121.
Saviozzi, A.; Levi-Minzi, R.; Cardelli, R.; Riffaldi, R. 2001. A comparison of soil quality in adjacent cultivated, forested, and native
grassland soils. Plant and Soil. 233: 251–259.
Tabatabai, M. A. 1994. Soil enzymes. In: Methods of soil analysis,
part 2, microbiological and biochemical properties. Madison, WI:
Soil Science Society of America: 775–833.
Tate, R. L. 1995. Soil microbiology. New York: John Wiley & Sons.
398 p.
53
Unique Characteristics of a
Systemic Fungal Endophyte
of Native Grasses in Arid
Southwestern Rangelands
Jerry R. Barrow
Abstract: Native grasses and shrubs in arid Southwestern rangelands are more extensively colonized by dark septate endophytic
fungi than by traditional mycorrhizal fungi. A histochemical method
was developed that revealed active internal structures that have
not been observed using conventional protocol for fungal analysis.
These fungi nonpathogenically and totally colonize all sieve elements, cortical and epidermal cells. They grow intercellularly and
intracellularly, forming an intimate, integrated association with all
root and leaf cells. Their internal morphology is often atypical of
commonly recognized symbiotic or pathogenic fungal colonization.
These endophytes accumulate and disperse large quantities of
lipids through the root. They form a saturated mucilage layer on
both the root and leaf surface, which protects them and maintains
hydraulic continuity with dry soil. Our results suggest that these
fungi function as carbon managers and enhance nutrient and water
uptake in arid ecosystems.
Mycorrhizal fungi profoundly influence plant communities in all ecosystems. The most extensively studied are the
arbuscular mycorrhiza (AM) that are associated with at
least 85 percent of all vascular plants. These fungi colonize
the root cortex, extend into the soil, and transport P to
arbuscules within the root cortex where it is released to the
plant. The term “mycorrhizae” means fungus root (Smith
and Read 1997). These fungi are harmonious components of
and are an extension of the root system. Fungi are well
adapted for nutrient acquisition; their small size allows
exploration of microscopic soil pores, and they are able to
function at lower water potentials than other organisms
(Griffin 1979). They function as microscopic pipelines where
carbon and minerals are actively transported simultaneously
to and away from the plant.
Little is known of the mycorrhizal status of desert plants,
and studies have focused primarily on the presence of
arbuscules in plant roots. Plant roots are also colonized by
many other kinds of fungi, including saprophytic or weakly
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Jerry R. Barrow is a Research Scientist at the USDA-ARS Jornada
Experimental Range, Las Cruces, NM 88003, U.S.A. FAX: (505) 646-5889,
e-mail: jbarrow@nmsu.edu
54
pathogenic fungi that may have symptomless endophytic or
biotrophic phases in their life cycles that are not apparent to
causal observers (Parbery 1996). Factors that distinguish
between beneficial and detrimental plant-fungus relationships are relative to the quantitative reciprocal exchange of
photosynthetic carbon for mineral nutrients, water, or protection (Smith and Smith 1996). Nonmycorrhizal plants are
frequently colonized by septate fungi that may function like
mycorrhizal fungi, but their study and importance have
been minimized because they do not conform to established
mycorrhizal morphology (Trappe 1981). Barrow and others
(1997) have found that dominant shrubs and grasses are
more extensively colonized by dark septate fungal endophytes (DSE) than by traditional mycorrhizae. DSE are
characterized by stained and melanized septate fungal structures (Jumpponen 2001). In a review by Jumpponen and
Trappe (1998), DSE fungi colonize a wide range of plant
species in stressed ecosystems. Barrow and Aaltonen (2001)
developed dual-staining methodology and high magnification differential interference microscopy that revealed substantially greater incidence of unique, active fungal structures that are not detected using conventional fungus-staining
methods and casual low-magnification observations.
Materials and Methods ___________
Material Collection
Roots and leaves were sampled weekly during 2001 from
a native population of black grama, Bouteloua eriopoda
(Torr.) Torr. (BOER), located on the USDA Agricultural
Research Service’s Jornada Experimental Range in southern New Mexico. Soil was chronically dry, and soil moisture
was generally less than 3 percent at most sampling times,
except for brief periods following precipitation events when
the soil was nearly saturated. Blue grama, Bouteloua gracilis (Kunth) Lag. Ex Steud. (BOGR), from various northern
and high elevations was also collected periodically. BOGR is
adapted to more mesic, higher elevations than BOER on the
Jornada Experimental Range. Some samples were taken
from actively growing sites that received normal and above
precipitation during the growing season. Fungal colonization
was determined to be consistent, and expression was uniform
in this and previous studies within the population at each
sampling period. For this study, roots from two to five plants
of each species were collected, bulked, and sealed in a plastic
bag and taken to the laboratory for preparation and analysis.
USDA Forest Service Proceedings RMRS-P-31. 2004
Unique Characteristics of a Systematic Fungal Endophyte of Native Grasses …
Tissue Preparation and Clearing
Methods developed by Bevege (1968), Brundrett and others (1983), Kormanik et al. (1980), and Phillips and Hayman
(1970) were modified for optimal visualization of fungi in
native grass roots. Roots were washed in tap water to remove
soil. From each bulked sample, healthy feeder roots of uniform maturity and appearance (approximately 0.250 mm in
diameter) were randomly selected and cleared by placing in
an autoclave in 2.5 percent KOH. Temperature was increased
to 121 ∞C over 5 minutes, maintained for 3 minutes, and after
8 minutes, samples were removed from the autoclave. Roots
were rinsed in tap water, bleached in 10 percent alkaline
H2O2 for 10 to 45 minutes to remove pigmentation, and placed
in 1 percent HCL for 3 minutes. Decolorized roots were rinsed
for 3 minutes in dH2O before staining with either trypan blue
(TB) or sudan IV (SIV), or they were dual stained with TB
and SIV. To stain with TB, roots were placed in prepared TB
stain (0.5 g trypan blue in 500 ml glycerol, 450 ml dH2O and
50 ml HCL), autoclaved at 121 ∞C for 3 minutes, and stored
in acidic glycerol (500 ml glycerol, 450 ml H20, and 50 ml
HCL). For SIV staining, roots were placed in prepared SIV
(3.0 g Sudan IV in 740 ml of 95 percent ETOH plus 240 ml
dH2O), autoclaved at 121 ∞C for 3 minutes, and stored in acidic
glycerol. For dual staining, roots were stained first in TB,
autoclaved as above, and destained 24 hours in the acidic
glycerol. Roots were then transferred to the SIV stain and
autoclaved as above. Dual stained roots were destained in
dH2O for 3 minutes, and roots were stored in acidic glycerol
until mounting.
Ten to twelve 2-cm root segments were placed on a microscope slide in several drops of permanent mounting medium.
A cover slip was placed over the root sections and pressed
firmly to facilitate analysis at high magnification. Analysis
was done with a Zeiss Axiophot microscope using both
conventional and DIC optics at 1000x. Leaves were prepared
similar to the roots.
Results ________________________
This method revealed a number of unexpected associations of DSE fungi with the native grama grasses that differ
from known pathogenic and symbiotic fungal associations
(Barrow 2003). Active fungal structures were dynamic and
morphologically different from typical fungal morphology
and were influenced by external environmental conditions.
The most prevalent structures in physiologically active roots
were fungal protoplasts that had no distinguishable walls.
As root activity decreased, structures with hyaline, stained,
or pigmented walls characteristic of DSE fungi became
increasingly evident and were connected with fungal protoplasts. A distinguishing feature of fungal structures was a
virtually constant association with lipid bodies of varied size
and shape. The fungus was systemic and formed an intimate, integrated interface with all root and leaf cells. The
nondestructive colonization of sieve elements of healthy
roots was unexpected. Colonization of cortical cells was less
evident in physiologically active roots, but as roots became
dormant, DSE fungi were observed in all cortical cells.
Another unique observation was the colonization of all root
meristematic cells. The fungus was observed extending from
USDA Forest Service Proceedings RMRS-P-31. 2004
Barrow
the root cortex and formed a loosely configured network
within a mucilaginous layer on the root surface. Fungi were
also observed extending from the root surface into the soil.
The same fungal morphology was observed associated with
sieve elements and bundle sheath and mesophyll cells of the
leaves. Colonization of epidermal cells and the stomatal
complex was likewise unique. Lipid bodies within fungal
structures that ranged from very small to occupying the
entire internal volume of the subsidiary cells were also
observed. Lipid-filled fungal protoplasts were also observed
extruding from the stomatal apertures.
Disscussion ____________________
The novel method used in this study revealed a substantially greater colonization of native grasses by DSE fungi
and atypical fungal structures that escape detection by
conventional methodology. Systemic colonization is assured
because of colonization of meristem cells and distribution of
the fungus at cell division to developing roots and shoots.
The nature and extent of fungal colonization indicate several potential ecological roles. The colonization of sieve
elements is presently relegated to pathogenic fungi and is
unique for beneficial or endophytic fungi. This also distinguishes DSE fungi from traditional mycorrhizal fungi that
are restricted to the root cortex and surface.
Staining with sudan IV, specific for lipids, revealed many
internal fungal structures that are not observed using conventional microscopy. The quantity and size of fungal lipid
bodies indicate substantial assimilation of host photosynthate. This would generally be viewed as a parasitic association; however, there is no detrimental affects to the host
plants. Therefore, it is concluded that DSE fungi are carbon
managers and utilize host carbon for the protection and
benefit of the host plant. Evidence for this is the protective
mucilaginous layer observed on both root and leaf surfaces.
Mucilage is an important organic carbon component of
plants and soil that affects water infiltration, retention, and
soil stability. This matrix is viewed as a protective, saturated microenvionment that would maintain hydraulic
continuity between the root and dry soil (Read and others
1999). Equally important is the colonization of cells of the
stomatal complex. Accumulation of lipids in the subsidiary
cells could physically influence stomatal closure. Lipid-filled
fungal protoplasts occupying stomatal apertures could potentially trap and retain transpired water. Such mechanisms would enhance water use efficiency and regulation
and plant survival in drought-stressed arid ecosystems.
References _____________________
Barrow, J. R. 2003. Atypical morphology of dark septate fungal root
endophytes of Bouteloua in arid Southwestern USA rangelands.
Mycorrhiza. 13: 239–247.
Barrow, J. R.; Aaltonen, R. E. 2001. A method of evaluating internal
colonization of Atriplex canescens (Pursh) Nutt. roots by dark
septate fungi and how they are influenced by host physiological
activity. Mycorrhiza. 11: 199–205.
Barrow, J. R.; Havstad, K. M.; McCaslin B. D. 1997. Fungal root
endophytes in fourwing saltbush, Atriplex canescens, on arid
rangelands of Southwestern USA. Arid Soil Research Rehabilitation. 11: 177–185.
55
Barrow
Unique Characteristics of a Systematic Fungal Endophyte of Native Grasses …
Bevege, D. I . 1968. A rapid technique for clearing tannins and
staining intact roots for detection of mycorrhizas caused by
Endogone spp., and some records of infection in Australian
plants. Transactions of the British Mycological Society. 51: 808–
810.
Brundrett, M. C.; Piche, Y.; Peterson, R. L. 1983. A new method for
observing the morphology of vesicular-arbuscular mycorrhizae.
Canadian Journal Botany. 62: 2128–2134.
Griffin, D. M. 1979. Water potential as a selective factor in the
microbial ecology of soils. In: Parr, J. F.; Gardner, W. R.; Elliot,
L. F., eds. Water potential relations in soil microbiology. SSSA
reprinted 1985, Special Publ. 9. Madison, WI: Soil Science Society
of America: 141–151.
Jumpponen, A. 2001. Dark septate endophytes: are they mycorrhizal? Mycorrhiza. 11: 207–211.
Jumpponen, A.; Trappe, J. M. 1998. Dark septate endophytes: a
review of facultative biotrophic root-colonizing fungi. New
Phytologist. 140: 295–310.
Kormanik, P. P.; Bryan, W. C.; Schultz, R. C. 1980. Procedures and
equipment for staining large numbers of plant root samples for
endomycorrhizal assay. Canadian Journal of Microbiology. 26:
536–538 .
Parbery, D. G. 1996. Trophism and the ecology of fungi associated
with plants. Biological Review. 71: 473–527.
Phillips, J. M.; Hayman, D. S. 1970. Improved procedures for
clearing roots and staining parasitic and vesicular-arbuscular
mycorrhizal fungi for rapid assessment of infection. Transactions
British Mycological Society. 55: 158–161.
Read, D. B.; Gregory, P.; Bell, A. E. 1999. Physical properties of
axenic maize root mucilage. Plant Soil. 211: 87–91.
Smith, S. E.; Read, D. J., eds. 1997. Mycorrhizal symbiosis. 2d ed.
San Diego and London: Academic Press: 59–60.
Smith, F. A.; Smith, S. E. 1996. Mutualism and parasitism: diversity in function and structure in the “arbuscular” (VA) mycorrhizal symbioses. Advances in Botanical Research: 22: 1–43.
Trappe, J. M. 1981. Mycorrhizae and productivity of arid and
semiarid rangelands. In: Manassah, J. T.; Briskey, E. J., eds.
Advances in food-producing systems for arid and semiarid lands.
Part A. New York: Academic Press: 581–593.
56
USDA Forest Service Proceedings RMRS-P-31. 2004
Cheatgrass Invasion Alters Soil
Morphology and Organic Matter
Dynamics in Big SagebrushSteppe Rangelands
Jay B. Norton
Thomas A. Monaco
Jeanette M. Norton
Douglas A. Johnson
Thomas A. Jones
Abstract: Cheatgrass (Bromus tectorum L.) is an invasive annual
grass that increases wildfire frequency, degrades native ecosystems, and threatens agriculture across vast areas of the Western
United States. This research examines how cheatgrass invasion
may alter physical and biological properties of soils. Proliferation of
very fine roots and high production of low-quality litter by cheatgrass increases porosity and near-surface microbial activity, which
may enhance decomposition of soil organic matter (SOM) similar to
cultivated systems. This may enlarge active SOM pools (mineral
and microbial biomass C and N) at the expense of slow pools and
humus. To test this hypothesis, soil properties beneath long-term
cheatgrass-invaded areas were compared with carefully matched
soils under shrub canopies and grass-covered interspaces at seven
undisturbed Wyoming big sagebrush (Artemisia tridentata ssp.
wyomingensis) plant communities in northern Utah and southeastern Idaho. Soils under cheatgrass had (1) higher porosity in surface
horizons, (2) higher concentrations of mineral N throughout the soil
profiles at the time of sampling, and (3) a larger proportion of
mineralizable C and N in total SOM of surface horizons than soil
under native shrub-steppe plant communities. These results support our hypothesis and suggest that long-term cheatgrass invasion
may alter ecological stability and resilience by depleting slow and
passive SOM pools. This research will contribute to improved
understanding of fundamental ecosystem processes required for
successful ecological restoration.
Introduction ___________________
The exotic annual grass cheatgrass (Bromus tectorum L.)
has invaded large areas of shrub-steppe ecosystems in the
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Jay B. Norton is County Director, University of California, Cooperative
Extension, Tuolumne County, Sonora, CA 95370; e-mail: jbnorton@
ucdavis.edu. Jeanette M. Norton is an Associate Professor, Department of
Plants, Soils, and Biometeorology, Utah State University, Logan, UT 843224820, U.S.A. Thomas A. Monaco, Douglas A. Johnson, and Thomas A. Jones
are Research Scientists, USDA Forage and Range Research Laboratory, 695
N 1100 E, Logan, UT 84322-6300, U.S.A. FAX: (435) 797-3075.
USDA Forest Service Proceedings RMRS-P-31. 2004
Western United States, and is now dominant on much of the
Great Basin, Snake River Plain, and Columbia Plain physiographic provinces (Knapp 1996; Knick 1999; Mack 1981).
The area covered by cheatgrass and its ecological impact on
shrub-steppe plant communities rivals ecosystem conversion
from grasslands to annual crops that took place in Central
North American (Samson and Knopf 1994). Numerous investigations of native ecosystem conversion to croplands
provide an understanding of agriculture-related environmental change, particularly of nutrient cycling and soil
organic matter (SOM) dynamics in both natural and disturbed ecosystems (Parton and others 1987; Schimel 1986).
As impacts of invasive plants on ecosystem processes begin
to be understood (D’Antonio and Vitousek 1992; Ehrenfeld
and Scott 2001), application of concepts from cropland conversion may further this understanding. In this paper, we
describe research that compared soils under cheatgrass to
those under native shrub-steppe grasslands. We discuss
ecological parallels between annual crops and cheatgrass
invasion and their implications for restoration of diverse
native plant communities.
On a worldwide basis, conversion of native ecosystems to
cropland results in about a 30-percent long-term decrease
in soil organic carbon (SOC). Annually, such conversion
accounts for nearly one-fifth of the net carbon (C) transfer
from terrestrial ecosystems to the atmosphere from changing land uses (Davidson and Ackerman 1993). In general,
cropland conversion is thought to change native ecosystems from net C sinks to important sources of atmospheric
CO2 (Schlesinger 1999). In semiarid shortgrass-steppe
grasslands, the long-term decrease in SOC from conversion
to cropland is often more than 60 percent (Aguilar and
others 1988; Schimel 1986), most of which is lost in the first
few years of cultivation (Bowman and others 1990). Cultivation also reduces labile C in soils as well as labile C as a
proportion of SOC (Bowman and others 1990). Cropland
soils typically have less total nitrogen (N), less labile N, and
lower labile N:total N than their uncultivated counterparts.
However, these N decreases with cultivation are proportionally less than that for SOC, resulting in narrowing C:N
ratios in both total and labile fractions after cultivation of
grassland soils (Bowman and others 1990). Inorganic N
57
Norton, Monaco, Norton, Johnson, and Jones
Cheatgrass Invasion Alters Soil Morphology and Organic Matter Dynamics in …
concentrations are often reported to be higher in cropland
than grassland soils, even without the effects of inorganic
fertilizers (DeLuca and Keeney 1993a). Native grasslands
are often described in terms of “tight,” conservative SOM
cycling, wherein diverse and growing microbial communities immobilize limited amounts of N as fast as it mineralizes from decomposing SOM (DeLuca and Keeney 1993a;
Schimel 1986). In contrast, cropland soils are thought to
have decaying microbial communities that “leak” inorganic
nutrients as SOM mineralization exceeds uptake of inorganic nutrients (Schimel 1986; Smith and others 1994).
Previous studies of cheatgrass invasion found greater concentrations of inorganic N in soils beneath cheatgrass than
native shrub-steppe vegetation (Evans and others 2001;
Svejcar and Sheley 2001). Other studies reported changes in
the amount of labile C as a proportion of TOC (Bolton and
others 1990; Gill and Burke 1999; Smith and others 1994).
Conversion from grassland to cropland changes SOM
dynamics in three ways: (1) reduced SOM inputs because of
crop harvest, (2) lower root:shoot ratios in annual crops
than perennial herbaceous and woody vegetation, and (3)
increased decomposition rates from aeration and more
labile organic substrates (Schimel 1986). Conversion of
native shrub-steppe to dominance by cheatgrass impacts
SOM dynamics in five similar ways: (1) reduced SOM
inputs from frequent fires that volatilize litter almost
annually in many areas (Knick and Rotenberry 1997;
Whisenant 1990), (2) reduced root:shoot ratios compared to
native grasses and shrubs (Monaco and others 2003), (3)
higher litter decomposition rates from more labile litter
NATIVE VEGETATION
compared to native plants, (4) altered soil structural properties (Gill and Burke 1999) due to differences in root
architecture between cheatgrass and native species
(Arredondo and Johnson 1999), and (5) different assemblages of soil microbes (Kuske and others 2002) and fauna
(Belnap and Phillips 2001) associated with cheatgrass and
native soils. Conversion from perennial vegetation to annual vegetation may lead to larger active (inorganic and
labile) SOM pools and smaller slow and passive (protected
SOM and humus) pools (Parton and others 1987) (fig. 1).
With cheatgrass invasion, these changes may result in
feedback processes that progressively alter soil structure
(Angers and Caron 1998), soil hydrology, and decomposition rates. Degradative feedbacks could ultimately lead to
impoverished soils that may limit restoration alternatives.
Similarities between the impacts of annual crops and
cheatgrass on SOM dynamics notwithstanding, clear differences in SOM content, composition, and cycling between
soils underneath cheatgrass-dominated and shrub-steppe
vegetation are difficult to detect (Bolton and others 1990;
Svejcar and Sheley 2001). The objective of our study was to
evaluate changes in soil morphology and the distribution and
composition of SOM associated with cheatgrass invasion of
sagebrush-steppe communities. Our underlying hypothesis
is that soils under cheatgrass-dominated vegetation exhibit
morphological characteristics and organic matter dynamics
that facilitate depletion of slow and passive SOM that turns
over in the timeframe of decades to centuries and enrich
active SOM that turns over every 2 years or less (fig. 1).
CHEATGRASS-DOMINATED
VEGETATION
STRUCTURAL
C
STRUCTURAL
C
METABOLIC
C
SHRUB-STEPPE
GRASSLAND
PLANT RESIDUE
METABOLIC
CHEATGRASSDOMINATED
PLANT RESIDUE
C
ACTIVE
SOIL
C
ACTIVE SOIL
C
SLOW
SOIL
C
SLOW
SOIL
C
PASSIVE
SOIL
C
PASSIVE
SOIL
C
Figure 1—Conceptual diagram of relative C flows in native shrub-steppe and cheatgrassdominated vegetation (based on Parton and others 1987). Structural and metabolic C in plant
residues have residence times of approximately 3 and 0.5 years, respectively. Active, slow, and
passive soil C pools have approximate turnover rates of 1.5, 25, and 1,000 years.
58
USDA Forest Service Proceedings RMRS-P-31. 2004
Cheatgrass Invasion Alters Soil Morphology and Organic Matter Dynamics in …
Materials and Methods ___________
Soil morphology and SOM dynamics were compared in soils
beneath cheatgrass-dominated and diverse Wyoming big sagebrush (Artemisia tridentata ssp. wyomingensis Beetle & A.W.
Young)-associated vegetation using a series of seven paired
sites in Utah and Idaho. Soil samples and data were collected during summer and fall 2001 from soil profiles and
replicated points randomly distributed within adjacent
cheatgrass-dominated and Wyoming big sagebrush-associated vegetation (fig. 2). Soil-forming factors (such as, parent
material, topography, climate, land use) at each paired site
were similar between cheatgrass-dominated and adjacent
sagebrush-steppe vegetation. Cheatgrass-dominated areas
were nearly monocultures with documented dates of conversion from native vegetation. Cheatgrass invasion was not a
result of livestock corrals, cultivation, excavation, or similar
forms of severe disturbance.
Soil profiles were described and sampled following soil
survey procedures (NRCS 1993). Soil profile descriptions
are reported in Norton and others (in press). Soil pits in
native vegetation were located with pit walls beneath grass
and shrubs representative for the area within each site. One
bulk soil sample was collected from three of the pit walls for
each soil horizon and placed on ice for transport to the
laboratory. Slope steepness, aspect, vegetation cover, description of landform and parent material, and likely cause
of cheatgrass invasion were recorded at each soil pit. Replicated samples were collected from two depths (0 to 5 cm and
5 to 20 cm) at five random bearings (0 to 360 degrees) and
distances (1 to 25 m) from each soil pit. Two distinct locations
were sampled in the native half of each pair, including an
area adjacent to the nearest perennial grass plants and an
area beneath the nearest shrub canopy (fig. 2). Soil samples
were homogenized shortly after collection, and about 10 g of
Study Design
Native
Cheatgrass
Norton, Monaco, Norton, Johnson, and Jones
each sample was placed in a preweighed sample cup that
contained 100 ml of 2M KCl for field extraction of nitrate-N
(NO3––N) and ammonium-N (NH4+–N). Cups were immediately capped and stored on ice for transport to the laboratory.
Vegetation at each site was evaluated for areal cover
(Daubenmire 1968) and frequency by species (Smith and
others 1987) along three transects, each with a total of 20
quadrats 0.25-m x 0.25-m square (fig. 2).
Immediately upon returning to the laboratory, field extractions were reweighed to determine exact amount of soil
sample, placed horizontally on a rotating shaker at 200 rpm
for 30 minutes, and then allowed to settle overnight in a 4 ∞C
refrigerator. Samples were filtered with Whatman no. 4
filters (Whatman International, Ltd., Maidstone, England)
and wet-sieved through 2-mm screens to remove gravel.
Gravel was dried, weighed, and the weight was subtracted
from the field-moist weight of the extracted sample. KCl
field extracts were frozen for further inorganic N analyses.
All soil samples were stored overnight in sealed sample bags
maintained at 4 ∞C in a refrigerator, sieved through 2-mm
screens (reserving at least three aggregates from each soil
profile sample for bulk density analysis), and partitioned for
determination of gravimetric moisture and mineralizable C
and N. Mineralizable C was determined by aerobic incubation of ~20-g samples brought to 23 percent soil water
content and incubated for 12 days in a 20 ∞C incubator in
0.95-L canning jars fitted with rubber septa. Gas in the
headspace was sampled on days 1, 6, and 12 during the
incubation (Zibilske 1994) and injected into a LI-COR 6400
infrared gas analyzer (LI-COR Corp., Lincoln, NE) for determination of CO2 concentration. Each jar was vacuum vented
and returned to the incubator after CO2 measurements on
day 1 and day 6. Inorganic N in the incubated ~20-g samples
was determined by 2M KCl extraction after the 12-day
incubation (Hart and others 1994). Concentration of NO3––
N and NH4+–N were determined with a Lachat flow injection
autoanalyzer (Lachat Instruments, Milwaukee, WI) for field
samples and postincubation samples. Postincubation inorganic N concentration (the amount of inorganic N after the
12-day incubation) represents the active N pool and is
reported as “active N.” Bulk density was determined by the
clod method (Blake and Hartge 1986). Total C and N were
determined by dry combustion with a Leco CHN 2000
Autoanalyzer (Leco Corp., St. Joseph, MI). Inorganic-C
concentration was determined gravimetrically (Loeppert
and Suarez 1996) and subtracted from total-C concentration
to determine organic-C concentration. Data from upper soil
horizons were analyzed by paired difference t-test and correlation procedures using the Microsoft Excel Data Analysis
Toolpak (Microsoft Corp., Redmond, WA). Cheatgrassnative comparisons for the native half of the replicated
samples were based on weighted means derived from percent
shrub cover and soil properties beneath grasses and shrubs.
Results and Discussion __________
Figure 2—Schematic diagram of study design.
Polygons represent shrubs, black dots represent
two-depth replicated soil samples, black rectangles
represent soil pits, and diagonal lines represent
vegetation transects. See text for further description.
USDA Forest Service Proceedings RMRS-P-31. 2004
Our results suggest that conversion of native shrubsteppe vegetation to cheatgrass dominance affects both the
soil environment and SOM dynamics in ways that are
analogous to grassland conversion to cropland. However,
absolute, whole-solum SOM loss is not as distinct with
59
Norton, Monaco, Norton, Johnson, and Jones
Cheatgrass Invasion Alters Soil Morphology and Organic Matter Dynamics in …
cheatgrass conversion as observed under cropland conversion
(Bowman and others 1990; Davidson and Ackerman 1993).
Our analysis of vegetation cover shows that the native
shrub-steppe areas were diverse, with up to 29 species in all
three life forms (grass, shrub, and forb) well represented at
the seven sites. Shrub cover ranged from 0 to 21 percent for
the native areas. In contrast, cheatgrass areas had a maximum of 18 species, many of which were exotic annual weeds.
Cheatgrass-dominated areas had very few shrubs, and most
sites had nearly 100 percent cheatgrass, although sandberg
bluegrass (Poa secunda J. Presl.) was well represented at
some sites (maximum of 13 percent cover). Thus, vegetation
at our sites was representative of cheatgrass-invaded areas
(Mack 1981).
Soil porosity (the inverse of bulk density) was significantly
higher in surface soil horizons under cheatgrass than native
shrub-steppe vegetation, but this difference diminished in
the subsurface (fig. 3). Higher porosity under cheatgrass is
likely the result of the dense, very fine, and shallow root mass
observed in A horizons under cheatgrass, which had 83 percent
more very fine roots (<1 mm diameter) than A horizons under
shrub-steppe vegetation. Fine-root density was the same
beneath the two vegetation types below the A horizons.
Cheatgrass soils had no roots coarser than 1 mm diameter,
while size classes of very fine through coarse were well
represented in shrub-steppe soils. Because cheatgrass dies
each summer, decaying roots leave behind a high density of
very fine tubular pores, which we observed in A horizons
under cheatgrass. This combination of greater porosity and
greater inputs of very fine roots accelerates decomposition
by enhancing air and water movement in surface soils, and
contributes to a relatively large and labile substrate each
year. These changes induced by cheatgrass may facilitate
pulses of microbial activity when moisture becomes available. These conditions may also enhance germination and
establishment of cheatgrass seedlings. However, this relatively high porosity and labile SOM content is dependent
on annual production by cheatgrass and may disintegrate
b
52
Percent
a
a
50
48
Native
46
Cheatgrass
44
a
42
40
38
Surface
Subsurface
Figure 3—Percent porosity of soils beneath cheatgrassdominated and native shrub-steppe vegetation based
on means of seven cheatgrass sites and seven native
sites. Different letters above columns indicate significant
differences (P< 0.05) resulting from paired-difference
analysis.
60
rapidly if cheatgrass germination is prevented, as with preemergent herbicides that often precede planting of perennial species.
Differences in SOM distribution and composition beneath
cheatgrass-dominated and shrub-steppe vegetation were
analogous to differences beneath grasslands and croplands,
but losses of SOM appeared to be less pronounced in shrubsteppe vegetation. The most significant differences occurred
in soil N fractions with higher concentrations of total N,
inorganic N, and active-pool N beneath cheatgrass than
shrub-steppe vegetation, and lower CO2–C:active-pool N
ratios beneath cheatgrass than shrub-steppe soils.
On a whole-solum basis, soils beneath cheatgrass had less
SOC and more total N than soils beneath shrub-steppe
vegetation at five of the seven sites (table 1). Although the
differences among averages for all sites were not significant (P > 0.05), cheatgrass solums had 30 percent less SOC
(P = 0.14) and 4 percent more N (P = 0.40) than shrub-steppe
solums. SOC concentrations in surface soils beneath
cheatgrass were equal to those under native shrub-steppe
vegetation (figs. 4 and 5), while total N concentrations were
higher in cheatgrass than shrub-steppe A horizons. This is
opposite to observed changes in grassland soils following
conversion to cropland (Bowman and others 1990). In our
study, SOC levels declined more with soil depth under
cheatgrass than under shrub-steppe vegetation at each of
the sites (fig. 4). Average SOC concentration in the upper
subsoil under cheatgrass averaged about half that of upper
subsoils under shrub-steppe vegetation. The magnitude of
the difference between SOC in A horizons and upper subsoils
under cheatgrass correlates with the approximate age of
cheatgrass invasion (fig. 6). Total N concentrations in the
subsoil were equal for the two vegetation types. Subsoils of
semiarid grasslands have higher decomposition rates than
surface horizons because they hold more moisture for more
of the year (Gill and others 1999). Higher soil water content,
along with lower root inputs to subsoils and the possibility
of a mineralization priming effect caused by large precipitation-induced pulses of microbial activity, may lead to depletion of subsoil SOC under cheatgrass-dominated vegetation.
A mineralization priming effect is often observed in agricultural soils when a labile C source is combined with aeration
caused by cultivation, which creates a large pulse of microbial activity. The C source is rapidly consumed, but enzymes
remain capable of mineralizing SOM stored in slow and
passive pools (DeLuca and Keeney 1993b). The result is
uptake of more SOM mineralization products than in the
original C source and depletion of SOM stored as slowly
decomposing materials and humus. Gill and Burke (1999)
found lower concentrations of slow-pool SOC beneath
cheatgrass than under adjacent shrub-steppe vegetation.
Field-extracted inorganic N differed significantly between
vegetation type and soil depth (table 2). Cheatgrass-dominated areas had 75 percent more (P = 0.02) NO3––N in the 0- to
5-cm soil depth and 66 percent more (P = 0.03) in the 5- to 20-cm
soil depth than shrub-steppe vegetation areas at the time of
sampling (table 2). Ammonium N concentrations (not presented) followed similar trends, but were highly variable.
Nitrate N also made up a higher proportion of total N at both
soil depths (table 2). Higher inorganic N concentrations are
often observed following conversion of grassland to cropland
USDA Forest Service Proceedings RMRS-P-31. 2004
Cheatgrass Invasion Alters Soil Morphology and Organic Matter Dynamics in …
Norton, Monaco, Norton, Johnson, and Jones
Table 1—Whole-solum organic C and total N under Wyoming big sagebrush steppe (native) and cheatgrass
vegetation (CG).
Site
Cedar Creek
Cedar Knoll
Five Mile Junction
Horse Butte
Hogup Mountains
Johnson Canyon
Mickel-Watson
Mean
b
0
8,639
Cedar Knoll
(Organic carbon concentration (g kg -1)
1
2
3
0
1
2
3
20
Depth (cm)
–30
Total N
Native
CG
Diff.
- - - - - g m–2 - - - percent
931
1,418
52
1,196
1,666
39
3422
2,328
–32
847
700
–17
1,076
1,230
14
502
627
25
1,079
1,480
37
P
0.14
1,293
1,350
4
P
0.40
Percent differences based on “native” values.
Calculated from bulk density of soil samples from each horizon.
Cedar Creek
0
12,299
40
60
80
35
R = 0.84
30
25
20
15
10
5
0
Native
0
Cheatgrass
Approximate age of cheatgrass invasion (years)
100
Figure 4—Soil organic carbon concentration in
paired soil profiles under cheatgrass-dominated
and Wyoming big sagebrush-steppe vegetation at
two of the seven study sites. SOC concentrations
followed similar trends at each of the seven study
sites.
48
b
46
44
42
a
40
38
Native
A-horizon—minimum organic carbon
(mg kg–1, top 50 cm)
a
Organic C
Native
CG
Diff.a
- - - - - - g m-2b - - - - percent
10,894
13,689
26
6,933
6,179
–11
36,625
15,135
–59
4,821
3,517
–27
11,187
6,282
–44
4,961
6,125
23
10,672
9,547
–11
Cheatgrass
Figure 5—Percent difference in SOC concentrations between the 0- to 5-cm soil depth and
the 5- to 20-cm soil depth based on means from
all replicated points. Different letters above
columns indicate significant differences (P < 0.05)
resulting from paired-difference analysis.
USDA Forest Service Proceedings RMRS-P-31. 2004
20
40
60
80
Figure 6—Difference in SOC concentration
between A-horizons (maximum in upper 50 cm)
and minimums within the top 50 cm in each of
seven cheatgrass soil profiles as a function of
approximate time since cheatgrass became the
dominant vegetation.
and are considered indicative of decaying, C-limited SOM
systems that “leak” mineralized nutrients after short-lived
pulses of microbial activity drive mineralization of stored
SOM (Schimel 1986). A similar explanation may apply to
higher inorganic-N concentrations beneath cheatgrass than
native shrub-steppe vegetation.
Active-pool C (C mineralized during 12-day incubations)
was nearly the same under the two vegetation types. Activepool N (postincubation inorganic N) concentrations, however, were 39 percent higher (P = 0.04) in the 0- to 5-cm soil
depth and 34 percent higher (P = 0.12) in the 5- to 20-cm soil
depth in cheatgrass-dominated areas than shrub-steppe
areas (table 2). Mineralized C accounted for about the same
proportion of total SOC in both vegetation types, but active
N accounted for a higher proportion of total N in cheatgrassdominated than shrub-steppe soils at both depth increments (table 2). These proportions suggest that cheatgrassdominated vegetation creates a mineralizing environment
61
Norton, Monaco, Norton, Johnson, and Jones
Cheatgrass Invasion Alters Soil Morphology and Organic Matter Dynamics in …
Table 2—Paired difference analysisa of selected soil properties under Wyoming sagebrush-teppe (native) and cheatgrass-dominated
vegetation (CG).
Nativeb
Depth
Organic C
CG
Diff.c
- - - g kg–1 - - 20.53
21.85
12.13
11.71
cm
0 to 5
5 to 20
Native
- - - g kg–1 - - percent
2.14
2.34
9.6
1.44
1.36 –6.0
percent
6.5
–3.5
Active-pool N
CG
Diff.
- - mg kg-1 - 23.71
33.05
5.60
7.52
0 to 5
5 to 20
Total N
CG
Diff.
Native
Field NO3––N
Native
CG
Diff.
- - mg kg–1 - 5.33
9.31
2.17
3.61
percent
74.8d
65.9d
Native
Active-pool C
CG
Diff.
- - - mg kg–1 - 190.48
210.31
64.33
60.50
percent
10.4
–6.0
NO3––N/Total N
Native
CG
Diff.
Active-pool N/Total N
Native
CG
Diff.
Active-pool C:N
Native
CG
Diff.
mg NO3––N g-1TN
2.54
4.07
1.53
2.92
mg NO3––N g-1TN
12.05
15.59
4.04
6.36
9.29
17.33
percent
39d
34
percent
60d
91d
percent
29e
57d
10.19
8.39
percent
10
–52d
a
Means from seven study sites, each with 10 native samples (five grass and five shrub) and five cheatgrass samples at each depth.
Weighted means based on proportion of grass and shrub cover on sites.
c
Percent difference based on “native” value.
d
Significant at the P < 0.05 level.
e
Significant at the P < 0.10 level.
b
where uptake is limited by shortage of labile C caused by
rapid decomposition and relatively low root:shoot ratios.
Ratios of SOC to total N were equal in soils of the two
vegetation types. The C:N ratios in the active SOM pool were
also equal in both A horizons and in the 0- to 5-cm depth of
both vegetation types (table 2). However, active-pool SOM of
cheatgrass soils had much lower C:N ratios in upper subsurface horizons (fig. 7) and in the 5- to 20-cm depth increment
(table 2) than those of native shrub-steppe soils. These
differences in active-pool C:N ratios are analogous to comparisons between grassland and cropland soils (DeLuca and
Keeney 1993a; Schimel 1986). This pattern suggests conservative, “tight” N cycling dominated by immobilization in
soils under shrub-steppe vegetation compared to more net
Cedar Creek
Cedar Knoll
Mineralized C:N after 12-d incubation
0
0
40
80
120
160
0
40
80
120
Depth (cm)
20
40
60
Native
160
mineralization in a “leaky” N cycle under cheatgrass-dominated vegetation. Smith and others (1994) found that soils
beneath shrub-steppe vegetation had much tighter, more
efficient N cycling compared to cropland, forest, and annual
grass ecosystems they analyzed.
Conclusions ____________________
This research contributes basic knowledge about how
extensive annual grass invasion alters soil properties recognized as foundations of ecological stability and resilience
(Coleman and others 1983). Our results suggest that conversion of diverse native plant communities to cheatgrass,
which is now the dominant vegetation on most of three major
physiographic provinces in the Western United States, leads
to losses of SOM and changes in its composition and distribution. These changes may be analogous to changes brought
about by crop cultivation in central grasslands of North
America, which are known to have altered basic ecosystem
functions such as cycling of C, N, and water. These basic
ecosystem functions are even more difficult to restore in the
arid and semiarid West because inherent fragility of altered
ecosystems (Tausch and others 1993) makes reversal uncertain and very expensive. The magnitudes of these changes
may intensify with the time that cheatgrass dominates,
limiting the prospects for restoring these areas. Our results
suggest that successful restoration of diverse native vegetation in cheatgrass-dominated areas may require use of
transition species that are capable of competing with
cheatgrass while contributing slow-pool SOC to the soil.
Cheatgrass
80
Figure 7—Active-pool C:N ratios (mineralized C:N after
12-d incubation) in paired soil profiles under cheatgrassdominated and Wyoming big sagebrush-steppe vegetation
at two of the seven study sites. Active-pool C:N ratios
followed similar trends at six of the seven study sites.
62
References _____________________
Aguilar, R.; Kelley, E. F.; Heil, R. D. 1988. Effects of cultivation of
soils in Northern Great Plains rangeland. Soil Science Society of
America Journal. 52: 1081–1085.
Angers, D. A.; Caron, J. 1998. Plant-induced changes in soil structure: processes and feedbacks. In: van Breeman, N., ed. Plantinduced soil changes: processes and feedbacks. Dordrecht, The
Netherlands: Kluwer Academic Publishers: 55–72.
USDA Forest Service Proceedings RMRS-P-31. 2004
Cheatgrass Invasion Alters Soil Morphology and Organic Matter Dynamics in …
Arredondo, J. T.; Johnson, D. A. 1999. Root architecture and
biomass allocation of three range grasses in response to nonuniform supply of nutrients and shoot defoliation. New Phytologist.
143: 373–385.
Belnap, J.; Phillips, S. L. 2001. Soil biota in an ungrazed grassland:
response to annual grass (Bromus tectorum) invasion. Ecological
Applications. 11: 1261–1275.
Blake, G. R.; Hartge, K. H. 1986. Bulk density. In: Klute, A., ed.
Methods of soil analysis. Part 1: physical and mineralogical
methods. 2d ed. Madison, WI: American Society of Agronomy:
363–375
Bolton, H., Jr.; Smith, J. L.; Wildung, R. E. 1990. Nitrogen mineralization potentials of shrub-steppe soils with different disturbance histories. Soil Science Society of America Journal. 54:
887–891.
Bowman, R. A.; Reeder, J. D.; Lober, R. W. 1990. Changes in soil
properties in a Central Plains rangeland soil after 3, 20, and 60
years of cultivation. Soil Science. 150: 851–857.
Coleman, D. C.; Reid, C. P. P.; Cole, C. V. 1983. Biological strategies
of nutrient cycling in soil systems. Advances in Ecological Research. 13: 1–55.
D’Antonio, C. M.; Vitousek, P. M. 1992. Biological invasions by
exotic grasses, the grass/fire cycle, and global change. Annual
Review of Ecology and Systematics. 23: 63–87.
Daubenmire, R. 1968. Plant communities: a textbook of plant
synecology. New York: Harper & Row, Publishers. 300 p.
Davidson, E. A.; Ackerman, I. L. 1993. Changes in soil carbon
inventories following cultivation of previously untilled soils.
Biogeochemistry. 20: 161–193.
DeLuca, T. H.; Keeney, D. R. 1993a. Soluble organics and extractable nitrogen in paired prairie and cultivated soils of central
Iowa. Soil Science. 155: 219–228.
DeLuca, T. H.; Keeney, D. R. 1993b. Glucose induced nitrate
assimilation in prairie and cultivated soils. Biogeochemistry. 21:
167–176.
Ehrenfeld, J. G.; Scott, N. 2001. Invasive species and the soil: effects
on organisms and ecosystem processes. Ecological Applications.
11: 1259–1260.
Evans, R. D.; Rimer, R.; Sperry, L.; Belnap, J. 2001. Exotic plant
invasion alters nitrogen dynamics in an arid grassland. Ecological Applications. 11: 1301–1310.
Gill, R. A.; Burke, I. C. 1999. Ecosystem consequences of plant life
form changes at three sites in the semiarid United States.
Oecologia. 121: 551–563.
Gill, R. A.; Burke, I. C.; Milchunas, D. G.; Lauenroth, W. K. 1999.
Relationship between root biomass and soil organic matter
pools in the shortgrass steppe of eastern Colorado. Ecosystems.
2: 226-236.
Hart, S. C.; Stark, J. M.; Davidson, E. A.; Firestone, M. K. 1994.
Nitrogen mineralization, immobilization, and nitrification. In:
Weaver, R. W.; Angle, S.; Bottomley, P., eds. Methods of soil
analysis. Part 2: microbiological and biochemical properties.
Madison, WI: Soil Science Society of America: 985–1018.
Knapp, P. A. 1996. Cheatgrass (Bromus tectorum L) dominance in
the Great Basin Desert. Global Environmental Change. 6: 37–52.
Knick, S. T. 1999. Requiem for a sagebrush ecosystem? Northwest
Science. 73: 53–57.
USDA Forest Service Proceedings RMRS-P-31. 2004
Norton, Monaco, Norton, Johnson, and Jones
Knick, S. T.; Rotenberry, J. T. 1997. Landscape characteristics of
disturbed shrubsteppe habitats in southwestern Idaho (U.S.A.).
Landscape Ecology. 12: 287–297.
Kuske, C. R.; Ticknor, L. O.; Miller, M. E.; Dunbar, J. M.; Davis, J. A.;
Barns, S. M.; Belnap, J. 2002. Comparison of soil bacterial
communities in rhizospheres of three plant species and the
interspaces in an arid grassland. Applied and Environmental
Microbiology. 68: 1854–1863.
Loeppert, R. H.; Suarez, D. L. 1996. Carbonate and gypsum. In:
Sparks, D. L., ed. Methods of soil analysis. Part 3: chemical
methods. Madison, WI: Soil Science Society of America: 437–474.
Mack, R. N. 1981. Invasion of Bromus tectorum L. into Western
North America: an ecological chronicle. Agro-Ecosystems. 7:
145–165.
Monaco, T. A.; Johnson, D. A.; Norton, J. M.; Jones, T. A.; Connors,
K. J.; Norton, J. B.; Redinbaugh, M. B. 2003. Contrasting responses of Intermountain West grasses to soil nitrogen. Journal
of Range Management. 56: 282–290.
Norton, J. B.; Monaco, T. A.; Norton, J. M.; Johnson, D. A.; Jones, T. A.
[In press]. Soil morphology and organic matter dynamics under
cheatgrass and sagebrush-steppe plant communities. Journal of
Arid Environments.
NRCS. 1993. Soil survey manual. Washington, DC: USDA Natural
Resource Conservation Service.
Parton, W. J.; Schimel, D. S.; Cole, C. V.; Ojima, D. S. 1987. Analysis
of factors controlling soil organic matter levels in Great Plains
grasslands. Soil Science Society of America Journal. 51: 1173–1179.
Samson, F. B.; Knopf, F. L. 1994. Prairie conservation in North
America. BioScience. 44: 418–421.
Schimel, D. S. 1986. Carbon and nitrogen turnover in adjacent
grassland and cropland ecosystems. Biogeochemistry. 2: 345–357.
Schlesinger, W. H. 1999. Carbon sequestration in soils. Science.
284: 2095.
Smith, J. L.; Halvorson, J. J.; Bolton, H. J. 1994. Spatial relationships of soil microbial biomass and C and N mineralization in a
semi-arid shrub-steppe ecosystem. Soil Biology Biochemistry.
26: 1151–1159.
Smith, S. D.; Bunting, S. C.; Hironaka, M. 1987. Evaluation of the
improvement in sensitivity of nested frequency plots to vegetational change by summation. Great Basin Naturalist. 47: 299–307.
Svejcar, T.; Sheley, R. 2001. Nitrogen dynamics in perennial- and
annual-dominated arid rangeland. Journal of Arid Environments. 47: 33–46.
Tausch, R. J.; Wigand, P. E.; Burkhardt, J. W. 1993. Viewpoint:
plant community thresholds, multiple steady states, and multiple successional pathways: Legacy of the Quaternary? Journal
of Range Management. 46: 439–447.
Whisenant, S. G. 1990. Changing fire frequencies on Idaho’s Snake
River Plains: ecological and management implications. In:
McArthur, D. E.; Romney, E. M.; Smith, S. D.; Tueller, P. T., eds.
Proceedings of the symposium on cheatgrass invasion, shrub dieoff, and other aspects of shrub biology and management; 1989
April 5–7; Las Vegas, NV. Gen. Tech. Rep. INT-GTR-276. Ogden,
UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station: 4–10.
Zibilske, L. M. 1994. Carbon mineralization. In: Weaver, R. W.;
Angle, S.; Bottomley, P., eds. Methods of soil analysis. Part 2:
microbiological and biochemical properties. Madison, WI: Soil
Science Society of America: 15–40.
63
Response of Lewis Flax
Seedlings to Inoculation With
Arbuscular Mycorrhizal Fungi
and Cyanobacteria
R. L. Pendleton
B. K. Pendleton
G. L. Howard
S. D. Warren
Abstract: Forbs comprise an important though understudied component of western rangelands. Little is known about the dependence of forb species on associated soil microorganisms such as
arbuscular mycorrhizal fungi. In 1996 and 1998, we conducted two
experiments examining the response of Lewis flax (Linum lewisii
Pursh) to inoculation with arbuscular mycorrhizal fungi under a
variety of soil conditions. Lewis flax was found to be highly dependent on mycorrhizae when grown in sandy soils, regardless of soil
nutrient levels. Mycorrhizal plants had 10 times the root biomass
and 16 times the shoot biomass of nonmycorrhizal controls. The
addition of mycorrhizae also significantly decreased root:shoot
ratios and significantly increased plant tissue concentrations of P,
K, and Zn. The addition of cyanobacterial inoculum to the soil
increased plant survival at low soil fertility, but had no effect on
seedling growth. In the second experiment, three accessions of flax
were grown in sand, peat, and a sand and peat mixture fertilized
with high levels of a slow-release fertilizer. All three flax accessions
responded positively to mycorrhizae in the sand and mixed soils,
growing five and three times larger, respectively. In contrast, no
significant response was observed in the peat soil. These findings
indicate that mycorrhizal dependence is complex, and can be influenced by a number of factors, including soil nutrient status and
organic matter content.
Introduction ___________________
Native forb species are prized both for their colorful
contribution to western landscapes and for their importance
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
R. L. Pendleton and B. K. Pendleton are Research Ecologists, USDA Forest
Service, Rocky Mountain Research Station, 333 Broadway SE, Suite 115,
Albuquerque, NM 87102-3497, U.S.A. FAX: (505) 724-3688, e-mail:
rpendleton@fs.fed.us. G. L. Howard is Research Biologist, U.S. Army Corps of
Engineers, Construction Engineering Research Laboratory, Champaign, IL.
S. D. Warren is Research Scientist, Center for Environmental Management
of Military Lands, Colorado State University, Fort Collins, CO 80523-1500.
64
in the diets of many herbivorous wildlife species. While early
revegetation efforts focused primarily on the use of grass
and shrub species, the use of herbaceous flowering plant
seed in revegetation mixes has become increasingly common. As a result, the establishment requirements of forbs
are becoming a research priority (Kitchen 1994).
North American Lewis flax (Linum lewisii) has wide distribution over the western half of the United States, Canada,
and Northern Mexico (Mosquin 1971). Lewis flax and its close
relative, L. perenne, have been used in revegetation and
restoration plantings and show horticultural potential in
xeriscaped gardens (Kitchen 1995; McArthur 1988). Other
flax species are known to benefit from mutualistic associations with arbuscular mycorrhizal fungi (Thingstrup and
others 1998). The purpose of this research was to examine
the relative response of Lewis flax seedlings to inoculation
with arbuscular mycorrhizal fungi (AMF) under a variety of
soil conditions, including inoculation with biological soil
crust organisms.
Methods ______________________
Experiment 1
Olympian 300XL pots (Hummert Int., Earth City, MO)
were filled with a low-fertility bank sand that had been
amended to one of three fertilizer levels. The sand was steam
sterilized at 65 ∞C for 1 hour to kill any indigenous mycorrhizal fungi that might be present, and 2.4 g ammonium
nitrate per 30 gallons of sand was added to raise the nitrogen
level to 10 ppm. Medium and high fertilizer levels were
amended with Osmocote 17-7-12 (The Scotts Company,
Marysville, OH), formulated for 12 to 14 months continuous
fertilization, applied at a rate of 5 and 9 ounces per cubic foot,
respectively. Each pot received one of four inoculation treatments: algal inoculation, mycorrhizal inoculation, dual
inoculation of both algae and mycorrhizae, and a
noninoculated control. Mycorrhizal inoculation consisted of
1 teaspoon (approximately 100 spores) of pot-cultured mycorrhizal inoculum applied 1 to 2 inches below the soil
surface. The inoculum was produced from soil collected from
USDA Forest Service Proceedings RMRS-P-31. 2004
Response of Lewis Flax Seedlings to Inoculation with Arbuscular Mycorrhizal …
Results _______________________
Experiment 1
Survival of experimental plants was influenced by soil
fertility level and algal inoculation (X2 = 47.8; P < 0.001).
Overall, survival was greatest at low fertility, followed by
medium fertility (fig. 1). Inoculation with algae significantly
2
improved survival at low fertility (X = 12.6; P < 0.001), but
2
significantly decreased survival at high fertility (X = 11.1;
P < 0.001).
Growth of flax plants was significantly affected by both
soil fertility and inoculation with mycorrhizal fungi (table 1).
30
Number of surviving plants
a mixed desert shrub community near Toquerville, UT, that
had been planted with ornamental corn seed (Carpenter
Seed Co., lot 125). Nonmycorrhizal pots received an equal
amount of inoculum that had been sterilized by autoclaving.
Soil crust inoculum consisted of pelletized algal inoculum of
the genera Microcoleus and Nostoc (Buttars and others
1998) ground to the consistency of cornmeal using a wheat
grinder and applied at a rate of 1.6 g per pot, sprinkled by
hand over the top of the soil at the time of planting.
Pots were sown with flax seed collected near Potosi Pass,
west of Las Vegas, NV. Flax plants were thinned to one
plant per pot upon emergence and grown for 15 weeks in a
thoroughly cleaned, temperature-controlled greenhouse. At
harvest, shoots were excised, dried, weighed, and ground for
analysis of mineral content. Tissue samples were analyzed
at the Soil and Plant Analysis Laboratory, Brigham Young
University, Provo, UT, using atomic absorption and microKjeldahl procedures (Johnson and Ulrich 1959). Results are
only reported for plants grown at high soil fertility, as
nonmycorrhizal plants at medium and low fertility produced
insufficient biomass for mineral analysis. Roots were washed
free of sand, dried at 60 ∞C, weighed, and stored at room
temperature pending examination of mycorrhizal colonization (Tarkalson and others 1998).
Data were analyzed using the General Linear Models
procedure on SAS, version 6.11 (SAS Institute 1993). Soil
fertility, algal treatment, and AMF treatment were used as
main effects in the model. Because of significant interaction
terms involving soil fertility, the effects of algal and AMF
inoculation treatments were also analyzed separately by soil
fertility level. Survival data were analyzed using a 3 x 2 x 2
contingency table, later subdivided by soil fertility level.
Pendleton, Pendleton, Howard, and Warren
+ algae
25
- algae
20
15
10
5
0
High
Experiment 2
Seeds of three accessions of native Lewis flax (Potosi Pass,
NV; Provo, UT; and Asotin, WA) were sown into SC-10 Super
Cells (Stuewe & Sons, Corvallis, OR) containing one of three
soil mixtures: (1) low-fertility bank sand that had been
steamed for 2 hours at 77 ∞C, (2) a peat-based commercial
potting soil, and (3) a mixture composed of half sand and half
commercial potting soil. Mycorrhizal inoculum of the species
Glomus intraradices, obtained from Tree of Life Nursery,
Capistrano, CA, was added to one-half of the cells approximately 2 inches below the soil line, at a rate of 1 tablespoon
per cell.
Fourteen replicate cells were used for each of the 18 soil/
accession/inoculum combinations. Plants were thinned to
one plant per cell and grown for 18 weeks in a temperaturecontrolled greenhouse. Each cell (plant) was fertilized using
1 tablespoon of Osmocote 17-7-12 (The Scotts Company,
Marysville, OH). Plant height was recorded at the end of 9,
14, and 18 weeks growth. After 18 weeks, shoots were
excised, dried, and weighed. Roots were washed free of soil,
dried at 60 ∞C, weighed, and stored at room temperature for
later examination of mycorrhizal colonization.
Data were analyzed using the General Linear Models
procedure on SAS, version 6.11 (SAS Institute 1993). Soil
type, plant accession, and AMF treatment were used as
main effects in the model. Data were also analyzed by soil
type and accession.
USDA Forest Service Proceedings RMRS-P-31. 2004
Medium
Low
Soil fertility level
Figure 1—Effect of algal treatment on survival of Lewis
flax plants (experiment 1). Bars marked with an asterisk
indicate significant differences (P £ 0.05) in plant
survival.
Table 1—Attained significance values for treatment effects on Lewis
flax growth measures by fertilizer level. Data are from
experiment 1.
Fertility/source
Shoot biomass
Root biomass
r:s ratio
0.0895
NEc
NEc
0.0025b
NEc
NEc
High fertility
AMF inoculation
Algal inoculation
AMF x algae
0.0135a
NEc
NEc
Medium fertility
AMF inoculation
Algal inoculation
AMF x algae
.0028b
.7954
.7169
.0043b
.8350
.8195
.0003b
.1068
.3659
Low fertility
AMF inoculation
Algal inoculation
AMF x algae
.0001b
.0588
.1386
.0001b
.1693
.2598
.1395
.0226a
.9970
Significant at P £ 0.05.
Significant at P £ 0.01.
c
NE = nonestimable due to missing data.
a
b
65
Pendleton, Pendleton, Howard, and Warren
Response of Lewis Flax Seedlings to Inoculation with Arbuscular Mycorrhizal …
Shoot biomass was significantly greater for mycorrhizal
plants at all soil fertility levels (fig. 2). Root biomass was also
significantly greater at medium and low soil fertility. Root:
shoot ratios increased with decreasing soil fertility, but were
consistently lower for mycorrhizal plants. Mycorrhizal plants
also had higher tissue concentrations of P, K, Zn, and Mn
5
Shoot biomass (g)
4
Experiment 2
AMF + Algae
AMF
Algae
Control
3
2
1
0
2
While flax accessions differed in inherent growth rate, all
accessions grew better with the addition of mycorrhizal
inoculum when grown in either the sand or sand and peat
mixture (fig. 3). Mycorrhizal plants were significantly taller
and had greater shoot and root biomass (table 3), as well as
total biomass, in these two soils. Mycorrhizal dependency
values for these soils ranged from 78 to 96 percent, indicating that Lewis flax is strongly dependent on the fungus. In
contrast, no significant differences in growth occurred in the
peat-based potting soil. Dependency values in the peat soil
ranged from 1 to 7 percent, indicating little or no dependence
under these growing conditions.
Conclusions ___________________
Root biomass (g)
1. Lewis flax may be characterized as mycorrhizal dependent. In inorganic media, growth of flax was significantly
enhanced by inoculation, having dependency values ranging
from 78 to 98 percent. The congener, L. usitatissimum, is
also mycorrhizal dependent, exhibiting increased growth,
reproduction, and resistance to pathogens in the presence of
the fungus (Dugassa and others 1996; Thingstrup and others 1998).
2. In contrast, dependency values in organic soil ranged
from 1 to 7 percent. The addition of organic matter through
the peat-based potting mixture in many ways mimicked the
response to P fertilization reported in previous studies.
Mycorrhizal dependency may be influenced by a number of
soil factors, including nutrient level, soil microorganisms,
and organic matter content (Hetrick and others 1986;
Schweiger and others 1995).
3. Inoculation with crust-forming algae increased survival at low soil fertility, but also increased root:shoot
ratios. A shift in biomass allocation toward increased root
mass suggests an increase in belowground competition for
soil resources. Some competition between plant roots and
soil microorganisms is likely, particularly during crust
establishment.
1.5
1
0.5
0
Root:shoot ratio
2
1
Acknowledgments _____________
0
High
Medium
Figure 2—Effect of inoculation treatments on
growth of Lewis flax plants (experiment 1).
66
(table 2). Calculated values for mycorrhizal dependence of
flax (Plenchette and others 1983) ranged from 72 to 98
percent.
Inoculation with blue-green algae increased root:shoot
ratios at low and medium fertility levels (table 1; fig. 2),
suggestive of some competition between plants and microorganisms. Algal inoculation also significantly increased plant
tissue concentrations of K (2.06 versus 1.45 percent; F = 9.2;
P = 0.01).
Low
This work was supported by grant MIPR W52EU260390969
from the U.S. Army Corps of Engineers, Construction Engineering Research Laboratory, Champaign, IL. Thanks are
extended to Bruce N. Smith of Brigham Young University
for providing student assistance, and to S. Garvin, D.
Tarkalson, and B. St. Clair for technical support.
USDA Forest Service Proceedings RMRS-P-31. 2004
Response of Lewis Flax Seedlings to Inoculation with Arbuscular Mycorrhizal …
Pendleton, Pendleton, Howard, and Warren
Table 2—Elemental tissue concentrations for mycorrhizal and nonmycorrhizal Lewis flax growing at high soil fertility. Nonmycorrhizal plants growing
at medium and low fertility yielded insufficient biomass for elemental analysis. Attained significance values (P) from statistical analyses are
given below. Significant differences are in boldface. Data are from experiment 1.
n
+ AMF
– AMF
P value
4
2
N
P
K
Ca
Mg
- - - - - - - - - - - - - - - - percent - - - - - - - - - - - - - - - - 4.16
0.33
2.01
2.73
0.69
3.85
.11
.91
3.07
.78
.3851
.0147
.0389
.5799
.4601
Zn
Fe
Mn
Cu
Na
- - - - - - - - - - - - - - - - - ppm - - - - - - - - - - - - - - - - - 17.47
203.1
67.3
10.83
743.5
4.43
280.3
128.5
10.47
740.3
.0310
.2994
.0082
.8881
.9922
0.6
+ AMF
Sand
- AMF
0.5
0.4
0.3
0.2
0.1
0
Total dry weight (g)
Sand:peat mixture
0.5
0.4
0.3
0.2
0.1
0
Peat-based medium
0.8
0.6
0.4
0.2
0
Asotin
Provo
Potosi Pass
Accession
Figure 3—Effect of mycorrhizal fungi on growth
of Lewis flax in three contrasting growth media
(experiment 2). Bars marked with an asterisk
indicate significant differences (P £ 0.05)
between mycorrhizal and nonmycorrhizal plants.
USDA Forest Service Proceedings RMRS-P-31. 2004
67
Pendleton, Pendleton, Howard, and Warren
Response of Lewis Flax Seedlings to Inoculation with Arbuscular Mycorrhizal …
Table 3—Attained significance values from ANOVAs on growth measures of three accessions of mycorrhizal and
nonmycorrhizal flax plants grown in sand, a peat-based potting soil, and a 1:1 mixture of the two. Data are from
experiment 2.
Soil/accession
Height
cm
Sand
Asotin
Provo
Potosi Pass
0.0001
.0001
.0001
Sand:peat
Asotin
Provo
Potosi Pass
Peat
Asotin
Provo
Potosi Pass
a
Shoot biomass
Total biomass
Root/shoot ratio
----------------- g ------------------0.0035
.0051
.0180
NS
.0043
.0076
0.0300
.0038
.0078
.0001
.0001
.0006
.0001
.0001
.0046
.0004
.0001
.0111
.0001
.0001
.0055
.0005
.0009
.0001
NSa
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
Not significant at the 0.05 level.
References ____________________
Buttars, S. M.; St. Clair, L. L.; Johansen, J. R.; Sray, J. C.; Payne,
M. C.; Webb, B. L.; Terry, R. E.; Pendleton, B. K.; Warren, S. D.
1998. Pelletized cyanobacterial soil amendments: laboratory testing for survival, escapability, and nitrogen fixation. Arid Soil
Research and Rehabilitation. 12: 165–178.
Dugassa, G. D.; von Alten, H.; Schönbeck, F. 1996. Effects of
arbuscular mycorrhiza (AM) on health of Linum usitatissimum L.
infected by fungal pathogens. Plant and Soil. 185: 173–182.
Hetrick, B. A. D.; Kitt, D. G.; Wilson, G. T. 1986. The influence of
phosphorus fertilization, drought, fungal species, and nonsterile
soil on mycorrhizal growth response in tall grass prairie plants.
Canadian Journal of Botany. 64: 1199–1203.
Johnson, C. M.; Ulrich, A. 1959. Analytical methods for use in plant
analysis. California Agricultural Experiment Station Bull. 766:
30–33.
Kitchen, S. G. 1994. Perennial forb life-history strategies on semiarid rangelands: implications for revegetation. In: Monsen, S. B.;
Kitchen, S. G., comps. Proceedings—ecology, management and
restoration of annual rangelands; 1992 May 18–22; Boise, ID.
Gen. Tech. Rep. INT-GTR-313. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain Research Station:
342–346.
Kitchen, S. G. 1995. Return of the native: a look at select accessions
of North American Lewis flax. In: Roundy, B. A.; McArthur, E. D.;
Haley, J. S.; Mann, D. K., comps. Proceedings: wildland shrub
and arid land restoration symposium; 1993 October 19–21; Las
68
Root biomass
Vegas, NV. Gen. Tech. Rep. INT-GTR-315. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Research Station: 321–326.
McArthur, E. D. 1988. New plant development in range management. In: Tueller, P. T., ed. Vegetation science applications for
rangeland analysis and management. Boston, MA: Kluwer Academic Publishers.
Mosquin, T. 1971. Biosystematic studies in the North American
species of Linum, section Adenolinum (Linaceae). Canadian
Journal of Botany. 49: 1379–1388.
Plenchette, C. J.; Fortin, A.; Furlan, V. 1983. Growth response of
several plant species to mycorrhiza in a soil of moderate P
fertility. I. Mycorrhizal dependency under field conditions. Plant
and Soil. 70: 191–209.
SAS Institute. 1993. SAS/STAT user’s guide, version 6, fourth
edition. Cary, NC: SAS Institute, Inc. 1686 p.
Schweiger, P. F.; Robson, A. D.; Barrow, N. J. 1995. Root hair length
determines beneficial effect of a Glomus species on shoot growth
of some pasture species. New Phytologist. 131: 247–254.
Talkalson, D. D.; Pendleton, R. L.; Jolley, V. D.; Robbins, C. W.;
Terry, R. E. 1998. Preparing and staining mycorrhizal structures in dry bean, sweet corn, and wheat using a block digester.
Communications in Soil Science and Plant Analysis. 29:
2263–2268.
Thingstrup, I.; Rubaek, G.; Sibbesen, E.; Jakobsen, I. 1998. Flax
(Linum usitatissimum L.) depends on arbuscular mycorrhizal
fungi for growth and P uptake at intermediate but not high soil P
levels in the field. Plant and Soil. 203: 37–46.
USDA Forest Service Proceedings RMRS-P-31. 2004
Heterogeneity in Soil C and N in
Late- and Mid-Seral SagebrushGrass Rangeland
J. D. Reeder
D. W. Grant
G. G. Jezile
R. D. Child
Abstract: Development of criteria and indicators that can be used
to evaluate nutrient cycling among plants, animals, and the soil is
essential for evaluating rangeland health and sustainability, as
well as for developing proper rangeland management practices. The
distribution of nutrients in space and time has been identified as a
useful criterion for evaluating the integrity of rangeland nutrient
cycles and energy flow in rangelands. Spatial distributions of soil
organic carbon (C) and organic and mineral nitrogen (N) were
measured in both mid- and late-seral stages of a sagebrush-grass
association. Vegetation cover differed between the two seral stages,
with grass cover significantly higher (15 versus 3 percent), and bare
ground area significantly lower (56 versus 69 percent) for the lateseral compared to mid-seral plant communities. The spatial distributions of soil organic C and N differed between the two seral stages
and corresponded to differences in vegetation cover. The proportion
of total soil N in the nitrate form was significantly higher in midseral soil than in late-seral soil, suggesting less efficient capture of
available N by the fragmented mid-seral plant community and an
uncoupling of the N cycle. Soil nitrate-N shows promise as a sensitive
indicator of seral stage in sagebrush-grass plant communities.
Introduction ___________________
In developing a recommended approach for evaluating the
ecological health of rangeland ecosystems, the National
Research Council (1994) recognized the fundamental importance of nutrient cycling to ecosystem health, and recommended that any comprehensive evaluation of rangeland
health and sustainability include indicators of the integrity
of nutrient cycles. They suggested that the best indicator of
rangeland nutrient cycling integrity is an assessment of the
distributions of nutrients in space and time. Soil organic
matter (SOM) is the primary source of plant-available nitrogen (N), the nutrient that most frequently limits primary
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
J. D. Reeder is a Soil Scientist and D. W. Grant a Biological Science
Technician, USDA-ARS, 1701 Centre Ave., Fort Collins, CO 80526, U.S.A.
FAX: (970) 492-7310, e-mail: jeanreeder@ars.usda.gov. G. G. Jezile was a
Graduate Student and D. W. Child is a Professor of the Department of
Rangeland Ecosystem Science, Colorado State University, Fort Collins, CO
80521, U.S.A.
USDA Forest Service Proceedings RMRS-P-31. 2004
and secondary productivity in rangeland ecosystems
(Woodmansee and others 1978). The distribution and turnover of SOM is therefore a major determinant of ecosystem
health and stability (Follett 2001).
Historically, rangeland studies have focused on the
aboveground components of the ecosystem, and relatively
few studies have evaluated belowground components and
processes. Basic knowledge is therefore limited concerning
the processes governing SOM distribution and turnover,
and the response of these processes to management or
disturbance. Additionally, detection of change in the mass or
distribution of SOM as an indication of change in rangeland
health is difficult because rangeland ecosystems are generally characterized by a high degree of spatial heterogeneity
in SOM (Burke and others 1999; Tongway and Ludwig
1994). The magnitude of change expected as the result of
disturbance or change in management is small relative to
the large SOM pool size and its spatial heterogeneity across
the landscape (Reeder 2002).
To address these limitations in our understanding of
rangeland nutrient cycling and change in cycle integrity in
response to management or disturbance, we compared soil
and vegetation characteristics of two seral stages of a
sagebrush-grass range to test the hypothesis that the
distributions and forms of SOM carbon (C) and N differ
between late- and mid-seral stages of a sagebrush-grass
plant community. The objectives were to (1) characterize
and quantify differences in soil C and N in late- and midseral stages of a Dry Mountain Loam sagebrush-grass
range; and (2) to identify forms of soil C and/or N that are
sensitive indicators of seral stage.
Methods and Materials __________
Study Site Description
The study was conducted on a Dry Mountain Loam Range
Site south of Walden, CO, at an elevation of 2,500 m on the
Cabin sandy loam soil series (fine loamy over sandy or
sandy-skeletal, mixed Argic Cryoboroll). Average annual
temperature is 2.8 ∞C, and average annual precipitation is
240 mm, with about two-thirds of annual precipitation
falling as rain during the growing season (NRCS 1981). The
Dry Mountain Loam Range Site consists of cool season
grasses and forbs that form a sparse stand beneath an open
69
Reeder, Grant, Jezile, and Child
stand of big sagebrush (Artemisia tridentata Nutt.). Estimated annual production ranges from 450 to 700 kg ha–1
(NRCS 1981).
Soil and vegetation sampling was conducted on two seral
stages (mid-seral and late-seral) with different management
histories. The areas selected for this study were based on the
range condition classification outlined by the Natural Resource
Conservation Service (NRCS 1981), and seral stages assigned
in 1997 by the Owl Mountain Partnership, Walden, CO.
Historic grazing management of the late-seral stage consisted of decades of heavy-to-moderate season-long grazing
prior to 1988, when the Bureau of Land Management implemented an allotment improvement program to improve
livestock distribution and provide rest periods from grazing.
Since 1988, grazing management has consisted of light-tomoderate grazing using a rotational grazing system that
limits annual grazing to 3 to 6 weeks per year, with the season
of grazing cycling annually among early season (beginning in
mid June), mid season (beginning in early to mid July), and
late season (beginning in mid August or early September).
Soil and vegetation sampling was conducted 12 years after
implementation of the new grazing management, at which
time the plant community had improved from “fair” to “good”
condition (NRCS 1981). At the time of sampling, late-seral
vegetation was composed of streambank wheatgrass (Agropyron dasystachyum), sheep fescue (Festuca ovina), junegrass
(Koeleria macrantha), sandberg bluegrass (Poa secunda),
muttongrass (Poa fendleriana), needlegrass (Stipa spp.),
squirretail (Sitanion hystrix Nutt.), bluebunch wheatgrass
(Pseudoroegneria spicata), big sagebrush, snake weed
(Gutierrezia sarothrae Pursh), pussytoes (Antennaria rosea),
stonecrop (Sedum lanceolata), hoods phlox (Phlox hoodii
Richards), moss phlox (Phlox bryoides), five fingers (Potentilla diversiflora), sedges (Carex spp.), lupine (Lupinus spp.),
false buckwheat (Eriogonum jamesii), clover (Trifolium spp.),
prickly gilia (Leptodactylon pungens), narrow leaved
mertensia (Mertensia lanceolata), and other forbs.
Mid-seral plant species were similar to those of the late
seral, but differed in abundance. Mid-seral vegetation consisted of increased big sagebrush, the warm season grass
blue grama (Bouteloua gracilis H.B.K., Lag. Ex Steud.), and
less palatable forbs, while cool season grasses were infrequent (NRCS 1981). Current and historic grazing management of the mid-seral stage consists of moderate-to-heavy
nonrotational grazing.
Field Plot Establishment and Sampling
Three 100-m transects were established in both seral
stages. Transect sites were selected on the basis of standard
soil classification techniques to assure that all transects
were positioned on the same soil series (Cabin sandy loam).
Species composition at peak production (late June) was
determined from visual estimates of ground cover from
twenty 0.1-m2 rectangular quadrats spaced 5 m apart along
each transect (Jezile and others, in press). The amounts of
bare ground, gravel, stones, ground litter, and standing
litter were recorded for each quadrat. Species cover was
expressed as a percentage of total cover, and estimates of
percentage composition of each species were used to calculate an NRCS range condition rating. Herbage yield was
estimated by randomly selecting and harvesting five of the
70
Heterogeneity in Soil C and N in Late- and Mid-Seral Sagebrush-Grass Rangeland
20 quadrats along each transect. For both seral stages,
clipped plant material was separated by dominant species,
dried for 48 hours at 60 ∞C and then reweighed to determine
the percentage dry weight of the dominant species (Jezile
and others, in press).
Soil samples were collected from two of the three transects
of both seral stages. Cores were obtained at the five randomly selected 0.1-m2 quadrat locations where herbage
yields had been sampled. At each coring location along each
transect, separate cores were obtained from three landscape
microsites: bare ground, within a grass patch, and under a
sagebrush canopy. Soil samples were taken with a 76.2-mmdiameter coring tube. Each soil core was partitioned into
three depth increments: 0 to 5, 5 to 10, and 10 to 20 cm.
Estimates of soil bulk density were obtained by the doublecylinder coring method described by Blake and Hartge
(1986) from three equally spaced locations along each
transect. Soil samples were placed in sealed plastic bags,
and transported in coolers to the laboratory for processing.
Laboratory Analyses
Soil samples were passed through a 2-mm screen to
remove plant crowns and visible roots. A subsample of fieldmoist soil was taken for analyses of gravimetric water
content and field-moist water soluble organic C (WSOC); the
remaining soil was air dried and used for all other analyses.
Total C and N were determined by dry combustion with a
Carlo Erba NA 1500 automatic carbon-nitrogen analyzer as
described by Schepers and others (1989). Inorganic C was
analyzed by a modified pressure-calcimeter method (Sherrod
and others 2002), and soil organic C was calculated by
subtracting inorganic C from total C. WSOC was determined
by the procedure of Davison and others (1987) with the
following modifications: the extraction ratio used was 5 mL
deionized water to 1 g soil, and filtrates were analyzed with
a Shimadzu TOC-5050A Soluble C analyzer. Exchangeable
nitrate (NO3–N) and ammonium (NH4–N) were extracted
with 2 M KCl and determined with an auto analyzer (Keeney
and Nelson 1982).
Statistical Methods
Differences in soil C and N among landscape microsites
(bare ground, beneath grass, and beneath shrub canopy)
were analyzed by depth and by cumulative depth with
analysis of variance and mean separation by least significant differences (LSD, P < 0.05) (Steele and Torrie 1980).
Differences in soil C and N between the two seral stages
were determined by depth and by cumulative depth with
paired t-tests (Steele and Torrie 1980).
Results _______________________
Vegetation
Based on estimates of plant species composition, calculated NRCS range condition ratings were 35 percent (“fair”
condition) for the mid-seral, and 57 percent (“good” condition)
for the late-seral stages (NRCS 1981). Jezile and others (in
press) reported that late-seral vegetation covered 44 percent
USDA Forest Service Proceedings RMRS-P-31. 2004
Heterogeneity in Soil C and N in Late- and Mid-Seral Sagebrush-Grass Rangeland
Soil organic C distribution
(0 to 20 cm)
100
80
Percent
of the soil surface (21 percent shrubs and canopy, 15 percent
grasses, and 7 percent forbs), while the remaining 56 percent
of the soil surface was comprised of exposed bare soil, gravel,
or litter-covered soil. In comparison, mid-seral vegetation
cover was 31 percent (23 percent shrubs, 3 percent grasses,
5 percent forbs), and 69 percent exposed soil, gravel, or littercovered soil. Mid-seral grass cover and total forage (grasses +
forbs) cover were significantly lower (P < 0.10), and bare
ground significantly higher, than in the late-seral stage
(Jezile and others, in press).
Reeder, Grant, Jezile, and Child
60
Bare
Grass
Shrub
40
20
Soil Properties
Masses of SOM-C and SOM-N in the late-seral stage were
significantly lower in bare areas than in grass or shrub
areas, whereas C and N in the mid-seral stage were not
significantly different among the three landscape microsites
(table 1). Spatial distribution of total SOM-C differed between the two seral stages, with significantly more C associated with bare-ground areas, and less associated with
grass areas, in the mid-seral than in the late-seral landscape
(fig. 1). Similar distribution patterns were observed for total
SOM-N (data not shown).
Labile fractions of soil C and N differed between the two
seral stages. WSOC was significantly lower, and nitrate-N
significantly higher, in the mid-seral compared to the lateseral landscape (table 1). Similar trends were observed in
the proportion of total SOM-C in water soluble forms (fig. 2),
and the proportion of total N in nitrate-N form (fig. 3). In
both the late- and mid-seral landscapes, mass of nitrate-N
was not significantly different among the three landscape
0
Late-seral
Mid-seral
Figure 1—Spatial distribution of soil organic C
in late- and mid-seral landscapes. (The two
seral stages differ (P < 0.01) in mass of soil C
in bare and grass areas.)
Ratio of soluble organic C to total soil organic C
(0 to 20 cm)
a
a
1.0
0.8
b
a
b
0.6
b
Late-seral
Mid-seral
0.4
Bare
Grass
Shrub
LSD (0.05)
Total soil organic C, Mg ha–1
Late-seral
Mid-seral
29.0
37.3
37.3
40.6
40.4
41.8
3.4
ns
pa
<0.01
.20
.30
Bare
Grass
Shrub
LSD (0.05)
Total soil organic N, Mg ha–1
Late-seral
Mid-seral
2.62
3.43
3.22
3.51
3.32
3.39
0.33
ns
pa
<0.01
.17
.37
–1
Bare
Grass
Shrub
LSD (0.05)
Bare
Grass
Shrub
LSD (0.05)
a
Water soluble organic C, kg ha
Late-seral
Mid-seral
2.16
1.84
3.36
2.28
3.94
3.02
.51
.33
Nitrate N, kg ha–1
Late-seral
Mid-seral
.98
3.10
.85
2.87
.97
2.72
ns
ns
pa
<0.01
<0.01
<0.01
pa
<0.01
<0.01
<0.01
0.2
0.0
Bare
Grass
Shrub
Figure 2—The ratio of mass of water soluble
organic C to mass of total organic C (x 104) in the
0- to 20-cm soil profile.
Ratio of Nitrate-N to total soil N
(0 to 20 cm)
1.0
NO3-N/total N (x 1000)
Table 1—Masses of carbon and nitrogen in the soil profile (0 to 20 cm)
of late- and mid-seral stages of a sagebrush-grass plant
community.
a
0.8
a
a
0.6
0.4
Late-seral
Mid-seral
b
b
b
Grass
Shrub
0.2
0.0
Bare
Figure 3—The ratio of mass of nitrate-N to mass
of total soil N (x 103) in the 0- to 20-cm soil profile.
Comparison of late- and mid-seral values (paired t-test).
USDA Forest Service Proceedings RMRS-P-31. 2004
71
Reeder, Grant, Jezile, and Child
microsites, but mass of WSOC was lowest in bare ground
areas, intermediate in grass areas, and highest beneath
shrub canopies (table 1).
Discussion ____________________
Jezile and others (in press) reported that SOM-C and
SOM-N concentrations were not significantly different between late- and mid-seral stages of this sagebrush-grass
plant community. When converted to a mass basis, we found
that levels of SOM-C and SOM-N were comparable between
late- and mid-seral landscapes in grass and shrub areas, but
levels of C and N were lower in bare-ground areas of the lateseral landscape than in the mid-seral landscape (table 1).
These data, plus visual field observations, suggest that the
late-seral landscape had experienced more historic soil loss
by erosion than had the mid-seral landscape.
Similar to the trends in C and N concentrations observed
by Jezile and others (in press), we found that total SOM-C
and SOM-N masses were not good indicators of seral stage.
Soil properties such as SOM are the result of millennia of
soil-forming processes (Jenny 1980). Thus, even though
significant changes in the plant community may be observed
in response to a particular management history spanning
decades, corresponding changes in most soil properties usually will not be observed unless erosional soil losses are
substantial (Lal and others 1999).
Although SOM consists primarily of a large stable pool of
C and nutrients that is resistant to decomposition, it also
includes small subpools of more readily decomposable or
labile organic compounds (Collins and others 1997) that may
be easily changed in response to perturbation such as grazing, or to grazing-induced changes in plant community
composition (Burke and others 1997; Reeder and others
2001). Assessing the integrity of nutrient cycles as an indicator of change in rangeland health (NRC 1994) can therefore be more readily accomplished by measuring the quality
rather than the quantity of SOM C and N in the soil. We
selected WSOC as an indicator of the labile pool of SOM-C
because it is a measure of root exudates that provide a
readily available C source for soil microorganisms (Cheng
and others 1993, 1996), and is strongly correlated with
microbial biomass and microbial activity (Burford and
Bremner 1975; Davidson and others 1987). We selected the
mineral forms of N in the soil, ammonium (NH4–N), and
nitrate (NO3–N), as indicators of the labile pool of soil N,
because these forms are easily measured, are the end products of the microbial mineralization processes that decompose SOM-N (Bartholomew 1965), and are the forms of soil
N that are available for direct uptake by plants (Harmsen
and Kolenbrander 1965).
We found that while comparisons of total masses of
SOM-C and N were not good indicators of seral stage and
thus of rangeland health, the proportions of total SOM-C
and N that were in labile forms clearly separated late-seral
from mid-seral landscapes in this sagebrush-grass plant
community. The proportion of total SOM-C that was soluble
was consistently higher (fig. 2), and the proportion of soil N
that was in the nitrate form consistently lower (fig. 3) in the
late-seral compared to the mid-seral landscape. These data
suggest that the N cycle is more tightly coupled in the late-seral
72
Heterogeneity in Soil C and N in Late- and Mid-Seral Sagebrush-Grass Rangeland
landscape. The higher level of WSOC observed in the lateseral soil suggests a higher level of root exudation and
associated increase in microbial activity to utilize the readily
available C (Cheng and others 1996). Stimulation of microbial activity by C-rich root exudates results in a N-limited
environment in which microorganisms immobilize soil N
(Kuzyakov 2002); thus levels of nitrate-N in the late-seral
soil are low. In the mid-seral landscape, higher levels of
nitrate-N and lower levels of WSOC indicate a loss of
integrity, or uncoupling, of the N cycle. The mid-seral landscape is characterized by a higher percentage of bare-ground
area and a more fragmented grass and forb plant community. In bare-ground and sparse-plant areas, warmer soil
surface temperatures in combination with pulsed rainfall
events enhance microbial decomposition of SOM (Bowman
and others 1990). In the absence of growing plants or in
areas of sparsely growing plants, root exudation of WSOC
compounds is low, so less stimulation of microbial activity by
C-rich root exudates occurs, resulting in net mineralization
rather than net immobilization of soil N, and nitrate-N
accumulation in the absence of sufficient plants to take up
excess mineral N.
Rangeland ecosystems continually respond to temporary
changes in both physical and biotic conditions. An assessment of change in rangeland health must be able to distinguish between changes that result in the crossing of an
ecological threshold from those that are temporary and the
result of normal fluctuations in physical and biotic conditions (NRC 1994). Our results demonstrate that assessments of labile forms of SOM, and of the landscape-level
spatial distribution of SOM-C and SOM-N, distinguished a
mid-seral from a late-seral stage of this sagebrush-grass
plant community. Measurement of soil nitrate shows promise as an indicator of the integrity of the N cycle and change
in successsional stage. Future studies are needed to evaluate the seasonality of differences in nitrate-N content between different seral stages, and to evaluate whether a
change in nitrate-N can serve as an early indicator of change
in rangeland health.
Acknowledgments _____________
We thank Ernest Taylor and Lucretia Sherrod for their
laboratory assistance with this research.
References ____________________
Bartholomew, W. V. 1965. Mineralization and immobilization of
nitrogen in the decomposition of plant and animal residues. In:
Bartholomew, W. V.; Clark, F. E., ed. Soil nitrogen. Madison WI:
American Society of Agronomy: 287–308.
Blake, G. R.; Hartge, K. H. 1986. Bulk density. In: Klute, Arnold, ed.
Methods of soil analysis. Part 1. Madison, WI: American Society
of Agronomy: 363–375.
Bowman, R. A.; Reeder, J. D.; Lober, R. W. 1990. Changes in soil
properties in a central plains rangeland soil after 3, 20 and 60
years of cultivation. Soil Science. 150(6): 851–857.
Burford, J. R.; Bremner, J. M. 1975. Relationships between the
denitrification capacities of soils and total, water-soluble and
readily decomposable soil organic matter. Soil Biology Biochemistry. 7: 389–394.
Burke, I. C.; Lauenroth, W. K.; Milchunas, D. G. 1997. Biogeochemistry of managed grasslands in central North America. In: Paul,
E. A.; Paustian, L.; Elliott, E. T.; Cole, C. V., ed. Soil organic
USDA Forest Service Proceedings RMRS-P-31. 2004
Heterogeneity in Soil C and N in Late- and Mid-Seral Sagebrush-Grass Rangeland
matter in temperate agroecosystems: long-term experiments in
North America. Boca Raton FL: CRC Press: 85–102.
Burke, I. C.; Lauenroth, W. K.; Riggle, R.; Brannen, P.; Madigan, B.;
Beard, S. 1999. Spatial variability of soil properties in the
shortgrass steppe: the relative importance of topography,
grazing, microsite and plant species in controlling spatial
patterns. Ecosystems. 2: 422–438.
Cheng, W.; Coleman, D. C.; Carroll, C. R.; Hofman, C. A. 1993. In
situ measurement of root respiration and soluble C concentrations in the rhizosphere. Soil Biology Biochemistry. 25(9):
1189–1196.
Cheng, W.; Zhang, Q.; Coleman, D. C.; Carroll, C. R.; Hoffman, C. A.
1996. Is available carbon limiting microbial respiration in the
rhizosphere? Soil Biology Biochemistry. 28(10/11): 1283–1288.
Collins, H. P.; Paul, E. A.; Paustian, K.; Elliott, E. T. 1997. Characterization of soil organic carbon relative to its stability and
turnover. In: Paul, E. A.; Paustian, L.; Elliott, E. T.; Cole, C. V.,
eds. Soil organic matter in temperate agroecosystems: longterm experiments in North America. Boca Raton, FL: CRC
Press: 51–72.
Davidson, E. A.; Galloway, L. F.; Strand, M. K. 1987. Assessing
available carbon: comparison of techniques across selected forest
soils. Communications in Soil Science and Plant Analysis. 18(1):
45–64.
Follett, R. F. 2001. Organic carbon pools in grazing land soils. In:
Follett, R. F.; Kimble, J. M.; Lal, R., eds. The potential of U.S.
grazing lands to sequester carbon and mitigate the greenhouse
effect. Boca Raton, FL: Lewis Publishers: 65–86.
Harmsen, G. W.; Kolenbrander, G. J. 1965. Soil inorganic nitrogen.
In: Bartholomew, W. V.; Clark, F. E., eds. Soil nitrogen. Madison
WI: American Society of Agronomy: 43–92.
Jenny, H. 1980. The soil resource. New York: Springer-Verlag. 377 p.
Jezile, G. G.; Reeder, J. D.; Grant, D. W.; Child, R. D. [In press]. Soil
and vegetation relationships on late- and mid-seral sagebrushgrass range.
Keeney, D. R.; Nelson, D. W. 1982. Nitrogen—inorganic forms. In:
Page, A. L., ed. Methods of soil analysis. Part 2. Chemical and
microbiological properties. Madison, WI: Americal Society of
Agronomy: 643–698.
USDA Forest Service Proceedings RMRS-P-31. 2004
Reeder, Grant, Jezile, and Child
Kuzyakov, Y. 2002. Review: factors affecting rhizosphere priming
effects. Journal of Plant Nutrition and Soil Science. 165: 382–396.
Lal, R.; Mokma, D.; Lowery, B. 1999. Relation between soil quality
and erosion. In: Lal, Rattan, ed. Soil quality and soil erosion. Boca
Raton, FL: CRC Press: 237–258.
National Research Council. 1994. Rangeland health: new methods
to classify, inventory and monitor rangelands. Washington, DC:
National Academy Press. 180 p.
Natural Resource Conservation Service. 1981. Soil survey of Jackson County Area, Colorado. U.S. Department of Agriculture,
Natural Resources Conservation Service and Colorado Agricultural Experiment Station. Washington, DC: U.S. Government
Printing Office. 159 p.
Reeder, J. D. 2002. Overcoming spatial variation in measuring soil
carbon stocks and sequestration potential of native rangelands in
the Western U.S. Proceedings of the OECD expert meeting on soil
organic carbon indicators for agricultural land; 2002 October 15–
18; Ottawa, Canada: 193–200.
Reeder, J. D.; Franks, C. D.; Milchunas, D. G. 2001. Root biomass
and microbial processes. In: Follett, R. F.; Kimble, J. M.; Lal, R.,
eds. The potential of U.S. grazing lands to sequester carbon and
mitigate the greenhouse effect. Boca Raton, FL: Lewis Publishers: 139–166.
Schepers, J. S.; Francis, D. D.; Thompson, M. T. 1989. Simultaneous
determination of total C, total N, and 15N on soil and plant
material. Communications in Soil Science and Plant Analysis. 20:
949–959.
Steele, R. G. D.; Torrie, J. H. 1980. Principles and procedures of
statistics: a biometrical approach. 2d ed. New York: McGrawHill. 593 p.
Sherrod, L. A.; Dunn, G.; Peterson, G. A.; Kolberg, R. L. 2002.
Inorganic carbon analysis by modified pressure-calcimeter method.
Soil Science Society of America Journal. 66: 299–305.
Tongway, D. J.; Ludwig, J. A. 1994. Small-scale resource heterogeneity in semi-arid landscapes. Pacific Conservation Biology. 1:
201–208.
Woodmansee, R. G.; Dodd, J. L.; Bowman, R. A.; Clark, F. E.;
Dickinson, C. E. 1978. Nitrogen budget of a shortgrass prairie
ecosystem. Oecologia. 34: 363–376.
73
Monitoring Rangeland Health:
Using a Biological Soil Crust
Stability Index
Roger Rosentreter
David J. Eldridge
Abstract: Monitoring of rangeland health has recently transitioned
from the taking of simple grass biomass measurements to a more
integrated plant biodiversity and soil surface condition evaluation.
One component being measured is the composition and cover of
biological soil crusts (bsc). The bsc cover has been shown to correlate
with soil stability. The bsc index is conducted in arid lands with
clumped vegetation along a 20-m line intercept. The bsc stability
index stratifies the bsc into one or more of three morphological
categories, records that category by cover class, and then calculates
a quantitative index. The bsc index is measured only in the plant
interspaces. This index allows for comparison with other sites and
for trend evaluation at the specific site over time. This bsc index
readily rates the stability of the plant interspace soil and requires
little analysis time or delay.
Keywords: monitoring, crusts, lichens, health, rangelands
Background ____________________
Biological soil crusts are integral components of rangeland soils on many continents. These complex assemblages
of lichens, liverworts, mosses, cyanobacteria, and algae
dominate the first few millimeters of the soil surface. Because of their close association with surface soils, crusts play
an important role in the regulation of water, sediment, and
nutrient flows, and provide favorable sites for the germination and establishment of some rangeland plants (Eldridge
and Rosentreter 1999; West 1990).
Biological soil crusts are generally regarded as indicative
of healthy landscapes due to the resistance they impart to
the soil surface against wind and water erosion. Despite this,
however, nonvascular plants have rarely been included in
broad-scale inventories or assessments of rangeland health
(West 1990). Biological soil crust organisms have traditionally been very difficult to identify, even to genus or family
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Roger Rosentreter is Botanist, U.S. Department of the Interior, Bureau of
Land Management, 1387 S. Vinnell Way, Boise, ID 83709, U.S.A. E-mail:
roger_rosentreter@blm.gov. David J. Eldridge is Senior Research Scientist,
Department of Sustainable Natural Resources, School of Biological, Earth
and Environmental Sciences, University of NSW, Sydney, 2052, NSW, Australia; e-mail: deldridge@unsw.edu.au
74
level, because of their small size, lack of definitive morphological features, and the fact that some species are separated
on the basis of their chemistry.
Techniques that use biological soil crusts to assess soil
health were pioneered in the semiarid woodlands of Eastern
Australia in the early 1980s (Tongway and Smith 1989).
Biological soil crust cover, along with other soil surface
features, provides an indication of the extent to which the
soil cycles nutrients, accepts rainfall, and resists erosion
(Tongway and Smith 1989). Despite the incorporation of
biological soil crusts into rangeland health assessment,
there has been a reluctance to use crusts as indicators due to
the problems discussed above. Eldridge and Rosentreter
(1999) proposed crust monitoring at the level of morphological groups, rather than using a species-based approach.
Morphological groups can be useful indicators of soil surface
stability due to the strong association between morphology
and function. This association provides an indication of the
degree to which single or groups of organisms can resist
perturbation.
Landscapes where the soil surface supports an extensive
cover of biological soil crusts are known to be relatively
resistant to wind and water erosion. Erosion, by removing
part of the soil surface horizon, is an indicator of a rangeland’s
declining health (Davenport and others 1998; Dormaar and
Willms 1998). Erosion reduces the productive potential of
the soil by decreasing soil organic matter, cation exchange
capacity and hence structural stability, and by reducing the
capacity of the soil to accept infiltration. Ultimately, depauperate soils have lower infiltration rates and excessive
overland flow, thereby reducing their structural stability
and production capacity.
Soil Crust Index _________________
A Biological Soil Crust Stability Index (BSCSI) integrating the surface cover of biological soil crusts with their
ability to protect the soil against erosion has been developed
for patterned grasslands and shrublands where cryptogamic crusts are a major component of the interspace soil
matrix. This index ranks morphological groups of cryptogamic taxa in terms of their increasing dimensionality and
hence their resistance to erosion. This is a practical system
of monitoring that simplifies and unifies previous methods
in order to encourage wider use and application. Even
morphological groups are sometimes difficult for observers
to determine during field survey. The soil crust index system
USDA Forest Service Proceedings RMRS-P-31. 2004
Monitoring Rangeland Health: Using a Biological Soil Crust Stability Index
Table 1—Dimensionality of biological soil crusts in relation to
morphological group and stability rating.
Crust dimenstion
type
1-D
2-D
3-D
Morphological groups
Stability rating
value
Cyanobacteria, fungi, and algae
Crustose, squamulose and
foliose lichens, and liverworts
Tall and short mosses, fructicose
lichens
USDA Forest Service Proceedings RMRS-P-31. 2004
An overall rating for a site is calculated by multiplying the
stability rating value for each crust type in each interspace
by a value that reflects its cover, and expressing that number
as a percentage out of 60, the maximum possible score. This
has the effect of relativizing sites and therefore enables
comparison between sites. The system is still under refinement and has yet to be tested in other environments.
Discussion _____________________
Although this system is still being tested and revised, it
appears to be of value regardless of the vegetation type as
long as the native vegetation is clumped or patterned
(Rosentreter and Eldridge 2002). Results to date have shown
strong relationships between the index and an independent
assessment of health of the rangelands (fig. 1). The Biological Soil Crust Stability Index may underestimate crust
contribution to total soil stability. However, this is a good
system for comparing sites over time in a relative sense. The
Index provides a rapid method of assessing the structure
and condition of interspaces in shrublands and grasslands.
This system should prove a useful tool for monitoring temporal and spatial changes in soil crust communities in
patchy environments.
The Biological Soil Crust Stability Index is an efficient
method because it requires less training, provides rapid and
statistically powerful data analyses, and allows for rapid
field measurements. Trampling impacts to the vegetation
and cryptogamic crusts are focused in the interspaces of
patterned arid vegetation. Therefore, impacts and changes
in range condition are first measurable in the interspaces
between plant clumps.
This Index evaluates microphytic vegetation and its impact without losing the small-scale structural features that
influence hydrologic and watershed stability. This system is
proposed as a rapid and efficient technique for monitoring
Index of landscape stability (%)
simplifies taxa even further according to their dimensionality, characterizing cyanobacteria and algae as one dimensional; crustose, squamulose and foliose lichens and liverworts as two dimensional; and fruticose lichens and mosses
as three dimensional.
Increasing dimensionality is considered indicative of
greater erosion control and increasing rangeland health.
One-dimensional soil crusts occur within the soil matrix and
may be difficult to see, but they resist raindrop impact and
hold the soil together, therefore protecting it from surface
waterflow and wind erosion. Their impact is most marked on
sandy soils where they often provide the only biological soil
protection. Two-dimensional soil crusts are similar in their
protection from erosion, but because they are generally
larger and grow on top of as well as within the soil, they
provide greater protection from raindrop impact, and the
small cracks between crust organisms may act as traps for
wind- and water-borne sediment. Three-dimensional soil
crusts are generally larger than two-dimensional crusts, and
their three-dimensional structure can capture and accumulate windblown dust and nutrients more effectively than
two-dimensional crust types.
In patterned vegetation communities in Idaho, U.S.A.,
and western NSW, Australia, a 20-m line-intercept method
has been used to determine the BSCSI. These measurements are often made in conjunction with other assessments
of rangeland health, which involve measurements of the size
and arrangement of vegetation patches.
The proportion of the various dimension types is multiplied by a rating value, which reflects the relative value of
the crust type to resist erosion, or deformation. Research is
currently underway to refine these stability ratings for
various dimensional types. For example, we are subjecting
the three crust types (1-D, 2-D, 3-D) to a range of laboratory
tests to examine their ability to resist stress (Emerson
aggregate test, water drop test), trap sediment (laboratory
simulations of microwind erosion), retain water (immersion
and drying tests), and withstand deformation (consistency
and coherency tests). Empirical data from these laboratory
tests will be subjected to multivariate analyses to determine
overall relative differences in effectiveness of the threedimensional types at stabilizing the soil surface. Preliminary results suggest that the 3-D surfaces are about 10 to 15
times more effective than 2-D surfaces, which are about 3 to
5 times more effective than 1-D surfaces (table 1).
Rosentreter and Eldridge
100
90
80
70
60
50
0
20
40
60
80
Biological soil crust index (%)
1
3 to 5
10 to 15
Figure 1—The relationship between the
Biological Soil Crust Stability Index (BSCSI)
and an index of landscape stability for
Artemisia shrublands in Idaho.
75
Rosentreter and Eldridge
erosion hazard potential, temporal and spatial changes in
the quality of plant and animal habitat, and wildfire risk.
References _____________________
Davenport, D. W.; Breshears, D. D.; Wilcox, B. P.; Allen, C. D. 1998.
Viewpoint: sustainability of pinyon-juniper ecosystems—unifying perspective of soil erosion thresholds. Journal of Range
Management. 51: 231–240.
Dormaar, J. F.; Willms, W. D.; 1998. Effect of forty-four years of
grazing on fescue grassland soils. Journal of Range Management.
5: 122–126.
76
Monitoring Rangeland Health: Using a Biological Soil Crust Stability Index
Eldridge, D. J.; Rosentreter, R. 1999. Morphological groups: a
framework for monitoring microphytic crusts in arid landscapes.
Journal of Arid Environments. 41: 11–25.
Rosentreter, R.; Eldridge, D. J. 2002. Monitoring biodiversity and
ecosystem function: grasslands, deserts and steppe. In: Nimis, P. L.;
Scheidegger, C.; Wolseley, P., eds. Monitoring with lichens—
monitoring lichens. NATO Science Series. The Hague, Netherlands: Kluwer Academic Publishers: 223–237.
Tongway, D. J.; Smith, E. L. 1989. Soil surface features as indicators
of rangeland site productivity. Australian Rangeland Journal.
11: 15–20.
West, N. E. 1990. Structure and function of microphytic soil crusts
in wildland ecosystems of arid and semi-arid regions. Advances in
Ecological Research. 20: 179–223.
USDA Forest Service Proceedings RMRS-P-31. 2004
Shrub Mounds Enhance Water
Flow in a Shrub-Steppe
Community in Southwestern
Idaho, U.S.A.
David J. Eldridge
Roger Rosentreter
Abstract: In the Western United States, substantial areas of
shrub-steppe have been replaced by annual exotic grasses during
the past century. These structural changes have been accompanied
by functional changes in the landscape in relation to the ability of
the landscape to capture essential resources such as water and
sediments. Surface waterflow was measured in three contrasting
shrub-steppe communities (sagebrush, winterfat, and rabbitbrush)
on mound and interspace surfaces using disc permeameters. Waterflow was substantially greater (2 to 6 times) under shrubs compared
with the shrub-free, microphytic-dominated interspaces. The ratio
of sorptivity under ponding to sorptivity under tension, which is a
useful index of the macropore (pores > 0.75 mm in diameter) status
of the soil, demonstrated that mounds were characterized by a
significantly greater amount of macropores compared with
interspaces. The dense cover of the moss Tortula ruralis, which
grows below Artemisia canopies, intercepted approximately 14 L of
water m–2 and enhanced the infiltration capacity of shrub mound
soils. The results reinforce the notion that these shrub-steppe
communities are partitioned into moisture-rich mounds separated
from moisture-deprived interspaces that are critical for the transfer
of water to the shrub mounds. Disturbance of the interspaces
ultimately leads to a reduction in water reaching the mounds and a
decline in shrubland function.
Introduction ____________________
Since the late nineteenth and early twentieth centuries,
extensive and uncontrolled livestock grazing has resulted in
substantial changes to the composition and structure of
many native shrub-steppe communities. The combined impact of grazing and invasion of annual Eurasian weeds
(Mack 1981), primarily cheatgrass (Bromus tectorum L.)
and medusahead (Taeniatherum caput-medusae ssp.
asperum L.), have increased the risk of wildfire in shrub
communities (Wambolt and others 2001). Consequently,
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
David J. Eldridge is Senior Research Scientist, Department of Sustainable Natural Resources, School of Biological, Earth and Environmental
Sciences, University of New South Wales, Sydney 2052, NSW, Australia,
Fax: 62 9385 1558, e-mail: d.eldridge@unsw.edu.au. Roger Rosentreter is
Botanist, Bureau of Land Management, 1387 S. Vinnell Way, Boise, ID
83709, U.S.A., e-mail: Roger_Rosentreter@blm.gov
USDA Forest Service Proceedings RMRS-P-31. 2004
increased frequency of wildfire has resulted in conversion of
substantial areas of Artemisia shrub-steppe to annual grasslands, which in turn burn at intervals of less than 10 years
(Whisenant 1990). This increasing cycle of fire, conversion to
annual grasslands, and more fire has increased unabated,
and now large areas of grassland exist in former shrubsteppe communities.
The dramatic change from shrublands to annual grasslands has resulted in substantial changes in landscape
function at both local and regional scales. Changes in landscape structure have resulted in major impacts on biodiversity, including loss of overstory and understory flora, declines in many faunal taxa, and invasion of exotic species
(Wambolt 2001, 2002; West 1993). Spatial heterogeneity in
many shrub-steppe communities worldwide is characterized by a matrix of microphytic crust-dominated soils (the
interspaces) on which are superimposed soil mounds colonized by perennial shrubs (shrub mounds). The microphytic
crusts of the interspaces comprise cyanobacteria, bacteria,
algae, mosses, and lichens (Hilty and others 2004; West
1990). Cyanobacterial polysaccharides bind together the
surface of the interspace soils, giving it a tight microstructure that generally repels water (Eldridge and others 2000).
In contrast, the surface soil of the shrub mound lacks the
microphytic crust and is covered with loose soil particles. On
functional shrub-steppe landscapes, these contrasting surfaces result in the concentration of water and nutrients
within the nutrient-enriched shrub patches at the expense
of the nutrient- and water-deprived interspaces (Blark and
Small 2003; Schlesinger and others 1990; Thiery and others
1995; Tongway and Ludwig 1994). This patchiness exists at
a range of spatial scales from whole catchments and
subcatchments to individual plants.
Efficient, sustainable production of arid shrublands is
dependent upon the movement of water, sediments, seed,
and soil from the interspace to the shrub mounds. When
these shrublands become degraded, resources are no longer
trapped within the shrub mounds, and the landscape becomes dysfunctional (Ludwig and Tongway 1995). Healthy
soils within shrub mounds are characterized by a high
degree of macroporosity, that is, a large number of pores,
generally biological in nature, greater than about 0.75 mm
in diameter. These pores are important for capturing excess
water generated from the interspaces. These macropores or
biopores are destroyed by trampling, disturbance, and cultivation. Waterflow through soils over small spatial scales can
77
Eldridge and Rosentreter
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community ...
be quantified using disc permeameters (White 1988). Using
the permeameters, a measure of the ratio of waterflow under
ponding to flow under tension can be obtained. This ratio
gives an indication of the relative health of the soil. Other
indicators of soil health include the relative differences in
infiltration between mound and interspace soils, as well as
the total infiltration capacity of the soil.
The aims of the research reported here are threefold.
Firstly, we wished to examine relative differences in infiltra∞ on three soil types supporting three different shrubtion
steppe communities common on the Snake River Plain.
Secondly, we wished to test whether there were significant
differences in infiltration between shrub mounds and adjoining interspaces in order to learn more about the consequences of shrub removal on the surface movement of water.
Thirdly, we were interested in examining the relative effect
of mosses growing below the shrub canopy on the interception, storage, and infiltration of water. We addressed these
questions by applying water to shrub mound and intershrub
(interspaces) areas using disc permeameters.
Methodology ___________________
Study Sites
Three sites were selected for detailed infiltration measurements: Bowns Creek (116∞18’00”W, 43∞12’00”N) approximately 48 k southeast of Boise, Cindercone Butte
(115∞56’30”W, 43∞20’30”N) approximately 56 km southeast
of Boise, and Kuna Butte in the Birds of Prey Reserve (116∞
25’32”N, 43∞25’59”W) approximately 20 km southwest of
Boise, ID. These sites were chosen because they contained
typical examples of three different types of shrub morphologies on different soils (loam, sand, and silt, respectively) at
sites close to Boise. Further, the sites were not adversely
impacted by encroachment of exotic annual grasses such as
cheatgrass and medusahead. All sites were on level (less
than 1 percent) slopes. A description of the biotic and abiotic
features of the sites is given in table 1.
Waterflow
The soil body consists of a matrix of soil, minerals, air, and
organic matter through which a network of pores of various
size and shape pass. Large soil pores (macropores), generally
greater than 0.75 to 1.00 mm in diameter, are important in
the transfer of water and nutrients through the soil (Bouma
1991) and are generally biological in origin. Macropores are
formed by plant roots and soil fauna (Oades 1993), and as
pore size increases, capillary tension declines and waterflow
increases. Smaller pores (micropores or matrix pores) are
also present between individual mineral grains and between
soil particles, but are not formed by soil biota.
The early phase of waterflow (2 to 10 minutes after water
is applied) is termed sorptivity, during which water enters
the soil in response to potential gradients of water potential
(influenced by soil dryness and pore structure), and gravitational potential (pore size, distribution and continuity; White
1988). The sorptivity phase is largely governed by the
capillary properties of the soil, particularly when it is dry. As
the soil wets up, gravitational forces become more important
(White 1988). The second phase of infiltration is known as
the steady-state phase. In uniform soils, a time is reached
where the flow rate from the source (in our case the disc
permeameter) stabilizes over time. This steady-state flow
rate or steady-state infiltration is governed by capillarity,
gravity, the area of the disc permeameter in contact with the
soil, and the pressure at which the water is supplied to the
soil surface (CSIRO 1988).
–0.5
Both sorptivity (mm h ) and steady-state infiltration
–1
(mm h ) were measured with disc permeameters at supply
potentials of –40 mm (tension) and +10 mm (ponded) (Perroux
and White 1988). When a negative pressure or tension is
applied to the soil using the disc permeameter, water is
prevented from entering macropores, and water only flows
through matrix pores. However, when infiltration is measured with a positive pressure, water flows through both
macropores and matrix pores. As these two tensions measure the contribution by different pore sizes, the ratio of
sorptivity at +10 mm to flow at –40 mm is a useful index of
the relative contribution of macropores to total waterflow.
This is an extremely informative measure, as macropores
are indicative of healthy, highly conductive soils, and a loss
of ecosystem function in terms of waterflow can be attributed in a large part to a loss of this macroporosity.
At each of the three sites, five replicate measurements of
waterflow were made at two microsites (shrub mound and
interspace). The ponded permeameter was placed on a steel
ring of 220 mm internal diameter, which was gently pressed
into the soil to a depth of about 7 to 10 mm, and sealed with
moistened soil along the outside edge to prevent leakage of
water. Both permeameters were placed alongside each other
and run for approximately 30 minutes, by which time steadystate had been achieved. At each supply potential, sorptivity
Table 1—Description of three sites.
Description
Bowns Creek
Cindercone Butte
Kuna Butte
Dominant vegetation
community
Wyoming big sagebrush
(Artemisia tridentata subsp.
wyomingensis)
Rabbitbrush
(Chrysothamnus nauseosus)
Winterfat (Krascheninnikovia
lanata) with scattered Wyoming
big sagebrush
Soil type
Loamy to clay loam Xerollic
Durargids and Xerollic
Haplargids
Loamy fine sand overlying
sandy loams; Xerollic
Camborthids
Silts and silty loams;
Durinodic Haplocalcid
Shrub density (ha–1)
7,140
81,600
30,200
Grazing status
Ungrazed
Intermittently grazed
Ungrazed exclosure
78
USDA Forest Service Proceedings RMRS-P-31. 2004
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community ...
Eldridge and Rosentreter
was calculated according to the method of Cook and Broeren
(1994), and steady-state infiltration according to White (1988).
60
a
Impact of Mosses on Infiltration
Soil Surface Morphology
At Bowns Creek, Cindercone Butte, and Kuna Butte we
measured the percentage cover of each of five surface morphology types (Hilty and others 2003) adjacent to each of the
five infiltration locations along two 5-m transects using the
line-intercept method (Canfield 1941). A detailed description of the five soil surface morphological types (coppice,
coppice bench, microplain, playette, disturbed), which have
been adapted from Eckert and others (1978), is given in Hilty
and others (2003).
Data Analyses
General Linear Models were used to test for differences in
sorptivity, steady-state infiltration, and soil surface morphology between the three sites and between the two
microsites and their interactions, after checking for homogeneity of variance (Levene’s test) using Minitab (1997). We
used a split-plot model with sites fixed and locations within
a site random. This enabled us to account for the variability
between the mound and nonmound microsites among the
five replicate locations at each site.
Results ________________________
Soil Surface Morphology
Overall, coppice, coppice bench, and microplain surface
morphologies accounted for about 90 percent of the surface of the soils across all sites. Although there were few
differences in the proportion of most surface cover types
between the three sites, there was a significantly greater
cover of coppice bench at Bowns Creek compared with the
other sites (F4,48 = 17.64, P < 0.001; fig. 1).
USDA Forest Service Proceedings RMRS-P-31. 2004
a
a
Cover (%)
Tortula ruralis, the dominant tall moss in sagebrush
steppe, is common in sheltered microsites beneath shrub
canopies (Rosentreter 1994) where it forms extensive mats
on the soil surface (Hilty and others 2003). In order to
measure the impact of this thick cover of moss on waterflow,
infiltration was remeasured, at the Bowns Creek site only,
at each of the exact same mound locations after all moss had
been carefully removed from the soil surface. We carefully
removed all of the moss without damaging the soil, then
allowed 5 days for the soil to equilibrate to its preinfiltration
level of soil moisture. Ponded infiltration was then remeasured over 25 minutes. The amount of infiltration into the
soil is the difference between the volume of water leaving the
permeameter and the volume of water intercepted and
stored by the moss. We removed the moss samples from
below the permeameters, transported them back to the
laboratory, and saturated them with water in order to
calculate the amount of water intercepted by mosses. Excess
water was allowed to drip off over a period of 20 minutes
before the samples were weighed.
Bowns Creek
Cindercone Butte
Kuna Butte
a
50
40
a
a
a
30
b
20
b
10
0
Coppice
Coppice
bench
Microplain
Playette
Trampled
Soil morphological class
Figure 1—Cover of the five morphological classes
at the three sites. Different letters within a class
indicate a significant difference in that class
between the three vegetation communities.
Infiltration
There were no significant differences in sorptivity or
steady-state infiltration under tension between the three
sites or between mound and interspace microsites (table 2;
fig. 2). Under ponded conditions, both sorptivity and steadystate infiltration were significantly greater at Cindercone
Butte compared with the other sites (F2,12 = 6.9 and 20.6,
respectively) (table 2; fig. 2). Both sorptivity and steadystate infiltration were significantly greater in soil below the
shrubs compared with in the interspaces (F1,12 = 28.1 and
22.7, respectively; P < 0.001). The greatest differences between shrub and interspace soils occurred at Bowns Creek
where steady-state infiltration under the shrubs was more
than 6.6 (+ 1.1 SEM) times that in the interspaces. Infiltration through the mounds at Cindercone Butte and Kuna
Butte was about twice that in the interspaces (fig. 2).
As indicated above, the ratio of sorptivity under ponding
to sorptivity under tension is a useful index of the macropore
status of the soil (White 1988). This ratio was significantly
greater on the shrub mounds (range 8.0 at Kuna Butte to
20.0 at Cindercone Butte) compared with the interspaces
(range 2.0 to 8.1; F1,12 = 16.8, P = 0.01), indicating the
abundance of macropores below the shrub mounds.
The rate of infiltration was generally more variable in the
presence of the moss Tortula ruralis. At about 17 minutes
after commencement of water application, infiltration in the
presence of moss cover significantly exceeded infiltration
with moss removed (fig. 3). Over the surface area of the
2
permeameter (220 cm ), Tortula ruralis intercepted a substantial amount of water (0.5 L), and by the cessation of
infiltration measurements, total infiltration of water in the
presence of moss (5.15 L) was 42 percent greater than that
without moss (2.99 L).
79
Eldridge and Rosentreter
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community ...
Table 2—Breakdown of site and microsite effects on sorptivity and steady-state infiltration under tension and ponding, and the
macropore status of soils.
Site effectsa
Parameter
Sorptivity under tension
Sorptivity under ponding
Steady-state infiltration under tension
Steady-state infiltration under ponding
Macropore statusc
Microsite effectsb
Trend
P
Trend
P
B=K=C
C > (B = K)
B=K=C
C > (B = K)
C > (B = K)
P = 0.19
P < 0.01
P = 0.79
P < 0.01
P = 0.04
M=I
M>I
M=I
M>I
M>I
P = 0.110
P < 0.001
P = 0.220
P < 0.001
P < 0.001
a
B = Bowns Creek, K = Kuna Butte, C = Cindercone Butte.
M = shrub mound, I = interspace.
c
Ratio of sorptivity under ponding (+10 mm) to sorptivity under tension (–40 mm).
b
Tension
50
a
a
400
a
25
a
b
200
b
b
0
0
200
50
a
mound
interspace
100
25
b
a
a
b
b
0
a
Ku
n
C
ns
Bu
re
e
tte
k
ne
co
Bo
w
er
in
d
a
Bu
re
e
Ku
n
ns
C
co
Bo
w
er
in
d
C
tte
k
ne
0
C
Steady-state
infiltration (mm h-1)
Sorptivity (mm h-0.5)
Ponding
Figure 2—Mean (+ standard error of the mean) sorptivity and steady-state infiltration
under ponding and tension on mound and interspace microsites at the three sites.
Different letters at a site indicate a significant difference at P < 0.05.
80
USDA Forest Service Proceedings RMRS-P-31. 2004
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community ...
Eldridge and Rosentreter
8
moss present
moss removed
Cumulative infiltration (L)
7
6
5
4
3
2
1
0
0
5
10
15
20
25
30
Time (min)
Figure 3—Mean (+ standard error of the mean)
cumulative infiltration (L) on five mounds with Tortula
ruralis intact and later removed. A significant
difference in cumulative infiltration occurs after
about 17 minutes.
Discussion _____________________
In our study, infiltration under the shrubs was two to six
times greater than infiltration in the interspaces (fig. 2).
We attribute the major difference in infiltration to structural differences in the two microsites, and in particular,
differences in the macroporosity status of each microsite.
Soil surface morphology was relatively similar between
sites (fig. 1), and higher sorptivity and infiltration at the
Cindercone site is attributed to the coarser (sandier) texture at that site (table 1).
Increased infiltration under shrubs has been demonstrated in a range of desert and semidesert environments.
Scholte (1989) showed that the rate of infiltration under
shrubs was 20 times greater than adjacent nonshrub surfaces, while greater flow under woody plants can be attributed to differences in macropores (Devitt and Smith 2002).
Greene (1992) demonstrated tenfold greater infiltration
rates in mulga (Acacia aneura) groves compared with adjacent sparsely populated runoff slopes. Our results emphasize the importance of shrubs for enhancing infiltration,
reinforcing the notion that in these environments shrubs act
as “fertile islands” (Parsons and others 2003).
A number of factors contribute to the different hydrologic
response near shrubs (Dunkerley 2000). Direct plant effects
include absorption of raindrop energy by the plant canopy
(Whitford and others 1997), which often changes soil water
levels and influences the microclimate of the surface, a
greater abundance of roots near the surface creating
macropores that facilitate waterflow (Beven and Germann
1982), and changes in soil litter cover and plant cover in the
USDA Forest Service Proceedings RMRS-P-31. 2004
vicinity of shrubs that influence the retention and absorption of water (Geedes and Dunkerley 1999). Shrubs can also
influence waterflow indirectly by altering adjacent physical
and chemical properties of the soil, making the soil more
conducive to soil invertebrates (Dunkerley 2000).
Despite the marked differences in sorptivity and steadystate infiltration under ponding, we failed to find any differences under tension when flow was restricted to pores
between individual soil particles (matrix pores). There were
no site effects nor any differences between shrub mounds
and interspaces (fig. 2). This indicates to us that the matrix
pore capacity of these soils is very similar, and that the soils
have a relatively similar inherent ability to conduct water,
which, in the absence of macropores, was quite low (fig. 2).
Thus, infiltration rates in these soils are largely driven by
macropores, and the greater infiltration under shrubs compared with the interspaces is due to the higher density of
biologically derived macropores, probably plant root and
faunal holes.
The movement of surface water is the critical process in
semiarid and arid landscapes where essential resources are
patchily distributed in the landscape. Some zones (source or
runoff zones) respond rapidly to rainfall by shedding runoff
water, directing it to adjoining patches which absorb water
(sink zones). The result is a series of fertile patches or
“islands” with enhanced soil moisture at levels greater than
that which they would normally receive through natural
rainfall (Yair 1994). In healthy landscapes these sinks cover
about 30 percent of the surface (Tongway 1990). The marked
differences in infiltration capacity between the shrub mounds
and interspaces reinforces our view that these shrublands
are strongly patterned into two distinct geomorphic zones:
(1) a water shedding interspace and (2) a water accreting
shrub mound (Ludwig and Tongway 1995).
On the sparsely vegetated interspaces, cryptogamic crusts
create a matrix of small runoff zones (Hilty and others 2003)
separated by the tussocks of small grasses, principally Poa
secunda, which function as localized sites for water accumulation. These Poa microhummocks, while contributing to
patchiness at small spatial scales, are thought to result from
previous periods of degradation, and may be a sign of a
recovering landscape (M. Pellant 2000, personal communication). During very small rainfall events, the Poa microhummocks hold most of the runoff water that is generated off
the relatively hydrophobic (water repelling) cryptogamic
crusts. However, larger rainfall and intense storms are
likely to generate runoff, which is captured in the shrub
mounds (Wainwright and others 1999) that have the capacity to absorb substantial volumes of runoff. Our infiltration
measurements indicate that the shrub mounds are capable
of soaking up substantial quantities of surface runoff, functioning as “ecological straws” in high rainfall years.
Some of this water retention in our study was due to the
extensive cover of the moss Tortula ruralis below the shrubs
(fig. 3). Tortula ruralis is generally restricted to shrub
canopy and grass microsites (Rosentreter 1984) where it
forms extensive mats on the soil surface. Tortula ruralis is
well adapted to prolonged desiccation (Longton 1992) and
quickly rehydrates after rainfall. Our study showed that
Tortula can store about 0.5 liters of water over an area of
2
–2
0.038 m , or 14.2 liters m . The destruction of moss cover
81
Eldridge and Rosentreter
after fire (Hilty and others 2004) has the capacity therefore
to reduce substantial amounts of water, which may be
important for invertebrates resident within shrub-steppe
ecosystems.
Shrub patches receiving enhanced water and nutrients
would be expected to have higher levels of soil nutrients,
particularly organic carbon and nitrogen, and greater populations of soil biota, and would be expected to be preferred
sites for germination and survival of vascular plants compared with the interspaces. The Chrysothamnus site at
Cindercone showed evidence of invasion by annual grasses,
particularly cheatgrass, and the shrub mounds were poorly
developed in comparison with the finer textured soils at
Kuna Butte and Bowns Creek (table 1). Nevertheless, landscape patchiness appears to be governed by the distribution
of shrubs, particularly Chrysothamnus and Artemisia spp.,
which show evidence of deep deposition of litter under the
canopy.
Disturbance of the well-developed microphytic crust in
the interspaces, particularly at Bowns Creek and Kuna
Butte, is likely to lead to increased infiltration in the
interspaces, invasion of the interspaces by annual species
such as cheatgrass, and reduced redistribution of runoff to
the shrub mounds. The breakdown in patchiness means that
annual rainfall alone (in the absence of runon) may be
insufficient to sustain the growth of perennial shrubs. Reduced infiltration through the mounds is likely to lead to a
change in the composition of vascular plants to one dominated by weedy ephemerals. In the short-term, death of the
perennial shrubs will result in reduced infiltration rates,
erosion around the mound base, reduced trapping of windand water-borne sediments (dust), and eventually partial
breakdown of the mounds (Offer and others 1998). The longterm effect of removal of the water-shedding crust is a
disintegration of the mounds, and a general decline in
rangelands productivity and condition (Eldridge and others
2000).
Conclusions ____________________
The results confirm the importance of landscape patchiness in controlling the distribution of water in shrub-steppe
ecosystems. Shrub mounds represent a stable, healthy landscape, and their enhanced water status results in higher
diversity and productivity. Activities that disturb the soil
surface in the interspace have the capacity to reduce runoff
to the shrub mounds and therefore reduce cover of the
shrubs. Long-term management of desert shrublands should
aim to maintain the integrity of both the mound and interspace, ensuring the efficient functioning of shrub and shrubland biota.
Acknowledgments ______________
We thank Cleve Davis (Bureau of Land Management) for
assistance with data collection, Terry Koen for statistical
advice, and Ann and Teddy DeBolt for comments on the
manuscript.
82
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community ...
References _____________________
Beven, K.; Germann, P. 1982. Macropores and waterflow in soils.
Water Resources Research. 18: l131–1325.
Blark, E. W.; Small, E. C. 2003. Associations between plant canopies
and the spatial patterns of infiltration in shrubland and grassland of the Chihuahuan Desert, New Mexico. Ecosystems. 6: 185–196.
Bouma, J. 1991. Influence of soil macropores on environmental
quality. Advances in Agronomy. 46: 1–37.
Canfield, R. H. 1941. Application of line interception in sampling
range vegetation. Journal of Forestry. 33: 388–394.
Cook, F. J.; Broeren, A. 1994. Six methods for determining sorptivity
and hydraulic conductivity with disc permeameters. Soil Science.
157: 2–11.
CSIRO. 1988. CSIRO disc permeameter instruction manual.
Canberra, ACT, Australia: CSIRO Division of Environmental
Mechanics.
Devitt, D. A.; Smith, S. D. 2002. Root channel macropores enhance
downward movement of water in a Mojave Desert ecosystem.
Journal of Arid Environments. 50: 99–108.
Dunkerley, D. L. 2000. Hydrological effects of dryland shrubs:
defining the spatial extent of modified soil water uptake rates at
an Australian desert site. Journal of Arid Environments. 45:
159–172.
Eckert, R. E.; Wood, M. K.; Blackburn, W. H.; Peterson, F. F.;
Stephens, J. L.; Meurisse, M. S. 1978. Effects of surface-soil
morphology on improvement and management of some arid and
semiarid rangelands. In: Hyder, D. N., ed. Proceedings of the
First International Rangelands Congress, Denver, CO: 299–302.
Eldridge, D. J.; Zaady, E.; Shachak, M. 2000. Infiltration through
three contrasting biological soil crusts in patterned landscapes in
the Negev, Israel. Catena. 40: 323–336.
Geedes, N.; Dunkerley, D. 1999. The influence of organic litter on
the erosive effects of raindrops and of gravity drops released from
desert shrubs. Catena. 36: 303–313.
Greene, R. S. B. 1992. Soil physical properties of three geomorphic
zones in a semi-arid mulga woodland. Australian Journal of Soil
Research. 30: 5569.
Hilty, J.; Eldridge, D. J.; Rosentreter, R.; Wicklow-Howard, M.
2003. Burning and seeding influence soil surface morphology in
an Artemisia shrubland in southern Idaho. Arid Land Research
and Management. 17: 111.
Hilty, J. H.; Eldridge, D. J.; Rosentreter, R.; Wicklow-Howard, M. C.;
Pellant, M. 2004. Recovery of biological soil crusts following wildfire in Idaho. Journal of Range Management. 57: 89–96.
Longton, R. E. 1992. The role of bryophytes and lichens in terrestrial
ecosystems. In: Bates, J. W.; Farmer, A. M., eds. Bryophytes and
lichens in a changing environment. London: Oxford University
Press: 32–76.
Ludwig, J.; Tongway, D. J. 1995. Spatial organisation of landscapes
and its function in semi-arid woodlands, Australia. Landscape
Ecology. 10: 51–63
Mack, R. N. 1981. Invasion of Bromus tectorum L. into Western
North America: an ecological chronicle. Agro-Ecosystems. 7: 145–
165.
Minitab. 1997. References manual, release 10.1. Pennsylvania, PA:
State College of Pennsylvania, Minitab Incorporated.
Oades, J. M. 1993. The role of biology in the formation, stabilization
and degradation of soil structure. Geoderma. 56: 377–400.
Offer, Z. Y.; Zaady E.; Shachak, M. 1998. Aeolian particle input to
the soil surface at the northern limit of the Negev Desert. Arid
Soil Research and Rehabilitation. 12: 55–62.
Parsons, A. J.; Wainwright, J.; Schlesinger, W. H.; Abrahams, A. D.
2003. The role of overland flow in sediment and nitrogen budgets
of mesquite dunefields, southern New Mexico. Journal of Arid
Environments. 53: 61–71.
Perroux, K. M.; White, I. 1988. Designs for disc permeameters. Soil
Science Society of America Journal. 52: 1205–1212.
Rosentreter, R. 1994. Displacement of rare plants by exotic grasses.
In: Monsen, S. B.; Kitchen, S. G., comps. Proceedings of the
ecology and management of annual rangelands. Gen. Tech. Rep.
INT-GTR-313, Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station: 170–175.
USDA Forest Service Proceedings RMRS-P-31. 2004
Shrub Mounds Enhance Water Flow in a Shrub-Steppe Community ...
Rosentreter, R. 1984. Compositional patterns within a rabbitbrush
(Chrysothamnus) community of the Idaho Snake River Plain. In:
McArthur, E. D.; Welch, B. L., comps. Proceedings—symposium
on the biology of Artemisia and Chrysothamnus. Gen. Tech. Rep.
INT-GTR-200. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station: 273–277.
Schlesinger, W. H.; Reynolds, J. E.; Cunningham, G. L.; Huenneke,
L. E.; Jarrell, W. M.; Virginia, R. A.; Whitford, W. G. 1990.
Biological feedbacks in global desertification. Science. 247:
1043-1048.
Scholte, T. S. 1989. Vegetation-soil relations in an area with sealed
Chromic Luvisols, Kenya. Arid Soil Research and Rehabilitation.
3: 337–348.
Thiery, J. M.; D’Herbes, J. M.; Valentin, C. A. 1995. A model
simulating the genesis of banded vegetation patterns in Niger.
Journal of Ecology. 83: 497–507.
Tongway, D. J.; Ludwig, J. A. 1994. Small-scale patch heterogeneity
in semi-arid landscapes. Pacific Conservation Biology. 1: 201–208.
Tongway, D. J. 1990. Soil and landscape processes in the restoration
of rangelands. Australian Rangeland Journal. 12: 54–57.
Wainwright, J.; Parsons, A. J.; Abrahams, A. D. 1999. Rainfall
energy under creasotebush. Journal of Arid Environments. 43:
111–120.
Wambolt, C. L.; Harp, A. J.; Welch, B. L.; Shaw, N.; Connelly, J. W.;
Reese, K. P.; Braun, C. E.; Klebenow, D. E.; McArthur, D.;
Thompson, J. G.; Torell, L. A.; Tanaka, J. A. 2002. Conservation
of greater sage-grouse on public lands in the Western U.S.:
implications of recovery and management policies. PACWPL
Policy Paper SG-02-02: 1–41.
USDA Forest Service Proceedings RMRS-P-31. 2004
Eldridge and Rosentreter
Wambolt, C.; Walhof, K.; Frisina, M. 2001. Recovery of big sagebrush communities after burning in south-western Montana.
Journal of Environmental Management. 61: 243–252.
West, N. E. 1990. Structure and function of microphytic soil crusts
in wildland ecosystems of arid and semi-arid regions. Advances in
Ecological Research. 20: 179–223.
West, N. E. 1993. Biodiversity of rangelands. Journal of Range
Management. 46: 2–13.
Whisenant, S. G. 1990. Changing fire frequencies on Idaho’s Snake
River Plains: ecological and management implications. In:
McArthur, E. D.; Romney, E. M.; Smith, S. D.; Tueller, P. T., comps.
Proceedings of a symposium on cheatgrass invasion, shrub die-off,
and other aspects of shrub biology and management. Gen. Tech.
Rep. INT-276. Ogden, UT: U.S Department of Agriculture, Forest
Service, Intermountain Research Station: 4–10.
White, I. 1988. Tillage practices and soil hydraulic properties: why
quantify the obvious? In: Loveday, J., ed. National Soil Conference Review Papers, Canberra, May 1988. Canberra, ACT, Australia: Society of Soil Science Incorporated: 87–126.
Whitford, W. G.; Anderson, J.; Rice, P. M. 1997. Stemflow contribution to the ‘fertile island’ effect in creosotebush, Larrea tridentata.
Journal of Arid Environments. 35: 451–457
Yair, A. 1994. The ambiguous impact of climate change at the desert
fringe: Northern Negev, Israel. In: Millington, A. C.; Pye, K., eds.
Environmental change in drylands: Biogeographical and Geomorphological Perspectives. London: John Wiley and Sons: 199–227.
83
Edaphic Characteristics of
Nitrogen Fixing Nodulation
(Actinorhizae) by Cercocarpus
montanus Raf. and Purshia
tridentata (Pursh) DC
Stephen E. Williams
Joe D. Johnson
Larry C. Munn
Terry Nieder
Abstract: Edaphology, excavation techniques, nodulation and some
nitrogen fixation characteristics of mountain mahogany (Cercocarpus montanus Raf.) and antelope bitterbrush (Purshia tridentata
(Pursh) DC) were examined principally in southeastern Wyoming
environments. Although critical levels of soil nitrogen and soil
organic matter were not identified, relatively high levels of nitrate
nitrogen (above 3 ug/g) and soil organic matter (above 2 to 4 percent)
inhibit nodulation, whereas low levels of nitrate (less than 1 ug/g)
and soil organic matter (less than 1 percent) are conducive to nodule
formation. Excavation of these nodulated Rosaceous plants in the
field requires care and attention to detail. Strategies for excavation
should involve allowing sufficient time and input of energy into the
excavation process, and excavation of mature plants may take even
weeks to accomplish. Field observations and subsequent greenhouse studies strongly suggest that the agents responsible for
nodulation of these plants (member of the actinomycete genus
Frankia) is either not present or is inactive in surface soils. The
paradox of stressing plants via sampling and the expectation that
collected data represents normal response is discussed.
Introduction ____________________
Cercocarpus montanus Raf. and Purshia tridentata (Pursh)
DC, mountain mahogany and antelope bitterbrush, respectively, are common shrubs in Western North America, being
found mostly in the foothills topographically between grasslands and forested environments. Sometimes both are in the
understory of lower elevation coniferous forest. These two
plants, both in the family Rosaceae, are important as browse
for many wildlife species and as cover for others, and both
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Stephen E. Williams and Larry C. Munn are Professors, Department of
Renewable Resources, College of Agriculture, University of Wyoming,
Laramie, WY 82071, U.S.A.; e-mail: sewms@uwyo.edu. Joe D. Johnson is
District Conservationist, USDA-NRCS, Kewaunee, WI 54216, U.S.A. Terry
Nieder is Technical Sales Representative, ADCON Telemetry, Santa Rosa,
CA 95403, U.S.A.
84
may periodically occupy deep soils, but mostly they can be
found in areas of shallow soil where the lithic contact is
within a meter of the surface. Usually the geological substrata, which best host both species, are sedimentary materials with limestone and sandstone; claystones and siltstone
are less favored.
That these two plants are adaptable to a diversity of
fairly harsh environments may well be a function of the
relatively large variety of mutualistic symbiotic microorganisms known to occupy the root systems of both plants.
Both plants are known nitrogen fixers, and are infected by
bacteria of the Actinomycetales genus Frankia. These
bacteria elicit development of root cortical hypertrophies
(nodules and often called actinorhizae) that function to
biologically fix atmospheric nitrogen. Further, both plants
can be infected by mycorrhizal fungi including arbuscular
mycorrhizal fungi expressed as an endomycorrhizal morphology as well as by fungi that are manifest as an
ectomycorrhizal morphology (Williams 1979). It is highly
probable that the capacity of these plants to host such a
variety of known beneficial root microorganisms results in
these plants being somewhat independent of essential soil
nitrogen because they can fix their own. They also can
benefit from the enhanced uptake function of mycorrhizal
roots, which includes uptake of critical nutrients such as
phosphorus and zinc (Allen 1991) as well as uptake of water
(Stahl and others 1998). Ectomycorrhizal plants are also
known to have some protection against root pathogens (for
example, Marx 1972).
Mountain mahogany and antelope bitterbrush (C. montanus and P. tridentata) are two species among perhaps a
dozen or more species spread across five or six genera in the
Rosaceae that fix nitrogen via association with Frankia. The
genera Chamaebatia, Cowania, Dryas, and perhaps Rubus
are also known for their nitrogen fixing habit (Peterson and
others 1991). It is not very well known how much nitrogen is
commonly fixed by these associations, although speculation
is that it is low especially in highly arid environments
(Dalton and Zobel 1977; Kummerow and others 1978). Further, there remains speculation that even though these
rosaceous plants have nodules and fix nitrogen that such
activity is not very important to supplying nitrogen needed
USDA Forest Service Proceedings RMRS-P-31. 2004
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
by the plant. This general observation is based on observations by a number of researchers that these plants are not
well nodulated or are often not nodulated in the field.
The objectives of this work are to (1) discuss soil properties
that influence actinorhizae development on these two plants,
(2) derive strategies for and show results of plant excavation
to locate and quantify actinorhizae development on these
two and similar plants, (3) report nodulation characteristics
and nitrogen fixation activity for these species under various
environmental conditions, and (4) discuss the uncertainty
that arises from determining normal characteristics of plant
activity when done destructively and where the plant is
otherwise stressed to secure such characteristics.
Materials and Methods ___________
Soil Properties That Influence
Actinorhizae Development
Mountain mahogany and antelope bitterbrush have been
excavated by the authors and associates since the mid-1970s
mostly in southeastern and central Wyoming. In the early
1980s to about 1995, numerous plants were excavated and
examined by excavation of the root system to a minimum
depth of 30 cm to consolidated and immovable rock or to the
extent of the root system. Plants of all ages were excavated;
however, the majority of plants were less than 12 years of
age. Here, excavation is used to mean that the entire root
system of each plant was systematically unearthed, nodules
counted, fresh and dry weights taken, and soils collected
from nodule zones. If a plant bore no nodules, soil was still
taken from that rooting zone of the plant where, from our
experience, we would normally expect to find nodules. This
zone was normally in the range of 30 to 60 cm from the soil
surface. Numerous plants were taken from highly disturbed
soils where top soil had been removed. We became aware
early that such plants were much more likely to be nodulated
than those on well developed soils.
Plants and soils were examined at approximately 50 sites
throughout Wyoming. Bitterbrush was examined primarily
in the Laramie Range, The Snowy Range foothills, the Wind
River Range foothills, and the Absaroka Range foothills.
Mountain mahogany samples were taken exclusively from
the Laramie Range and the Snowy Range foothills.
Plants were examined visually for presence of nodules.
Organic matter (Allison 1965), NO3-N (Oien and Olsen
1969), pH (1:1 in water), available K (Bower and Wilcox
1965), available P (Wantanabe and Olsen 1965), and electrical conductivity (Bower and Wilcox 1965) were determined
for each soil sample.
Strategies for Plant Excavation
Techniques—The tendency when looking for nodules on
root systems is to take digging tools to the field and oftentimes
indiscriminately and randomly sample from the soil around
the base of the plant, relying mostly on chance that one will
strike a cluster of nodules. This rather arbitrary method
can be successful when nodules are needed for general
examination or for a crude assay. It is particularly fruitful
when looking for nodules on cultivated legumes, and can be
USDA Forest Service Proceedings RMRS-P-31. 2004
Williams, Johnson, Munn, and Nieder
successful when examining members of the genus Alnus in
many riparian systems. However, when nodules are deep
within soils, and when clusters are apparently rare, more
methodical methods are used, and sufficient time must be
allocated to fully excavate plant roots.
There are numerous problems encountered when quantitatively examining plant root systems. Actinomycete nodulated shrubs often have root systems that extend to considerable depth, the rooting medium may be obstreperous (for
example, too clayey, too many large coarse fragments), roots
of differing species make following the roots of the target
individual difficult, roots of same species make following
target individuals difficult, and plants can be in steep
topographical positions that may defy examination.
We have used a variety of techniques to examine roots
systems of shrubs. Herein we describe methodology for
shrub excavation that can be used to quantitatively remove
shrub root systems. The objectives and scale of research will
dictate, to some degree, the intensity of excavation needed
and the data that will be recorded. However, as the spatial
and temporal development of plant roots are examined, as
well as symbiotic associations between roots and a variety of
mutualistic, pathogenic, and commensalic biota, roots systems will need to be more accurately and methodically
excavated. There are a variety of reference materials that
can be accessed that detail root excavation considerations
and techniques. Two works from the archaeology literature
are very much worth consideration especially for project
level excavations (Joukowsky 1980; Roskams 2001).
Examination of Soil Taxa Associated With Nodulated Shrub Communities—During plant excavation to
locate and quantify actinorhizae development, soils and
plants were examined along a trenched transect that was
made from a site occupied by mountain mahogany, through
the boundary of the plant community, and into an associated
soil type having no mountain mahogany and dominated
primarily by grasses and forbs. The site was representative
of a number of associations between mountain mahogany
and adjacent vegetation and was located 10 km NNE of
Laramie, WY, south of the mouth of Roger Canyon.
Soils and plants were examined also along a transect
made at a site occupied by antelope bitterbrush, through the
boundary of the antelope bitterbrush community, and into an
associated soil type that was void of the shrub but dominated
primarily by Artemisia sp. shrubs, grasses, and forbs. The
transect was located 15 km ESE of Laramie, WY, adjacent to
Blair Road. This transect was not trenched, but rather was
examined by locating pits and observation of natural and
manmade cuts that intersected the established transect.
The two transects indicated above were established primarily to determine if there were identifiable morphological
features of soils associated with these two rosaceous shrub
communities. Soils were classified in accordance with Soil
Taxonomy (Soil Survey Staff 1999) and the Soil Survey of
Albany County, WY (Reckner 1998).
Nodulation Characteristics and Nitrogen
Fixation Activity
Actinorhizal Plants During Winter Months—Antelope bitterbrush and mountain mahogany were observed
85
Williams, Johnson, Munn, and Nieder
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
occasionally during winter seasons between 1988 and 1998.
Observation years were categorized as dry winters or wet
winters where dry winters had very little surface snow
accumulation and the soil profile was generally dry. Wet
winters were those where there was considerable surface
accumulation of snow.
During one wet winter, 1988, antelope bitterbrush individuals were excavated from a site along the Happy Jack
Road on Pole Mountain near Laramie, WY. Soils here were
weakly developed with some organic matter accumulations
in the upper 20 to 30 cm over an undeveloped coarse subsoil.
Nodules were found in the coarse subsoil. Three plants were
excavated in February of that year—a 1-year-old plant on a
recently disturbed roadcut and two 5-year-old plants on an
adjacent, but undisturbed site. Nodules were taken and
incubated in a 10 percent atmosphere of acetylene to ascertain nodule nitrogen fixation via analysis of ethylene
(Upchurch 1987). Incubation chambers were 450 ml volume,
incubation time was up to 1 hour, and was done at ambient
soil temperatures in the field. Gas samples from the incubation chambers were stored in vacutainers prior to analysis.
Subsamples of 100 microliters each were analyzed via gas
chromatography (Upchurch 1987). Observations of antelope
bitterbrush and mountain mahogany during several winter
seasons showed that during dry winters, nodules were very
hard to find and that when found, they had no visual signs
of activity. Nodules were usually brown to black including
nodule tips. During wet winters and especially under snow
pack, nodules were easier to find and often gave the visual
appearance of considerable activity, that is, nodule tips were
white and apparently turgid.
Biological Assay of Soil Materials for Nodulation
Capacity—The field portion of this study was done in SE
Wyoming, in Laramie County. Seven sites were located in
two separate areas. Six of these sites were inhabited by
mountain mahogany, and one site that did not have this
plant provided soil materials as a control. Pits were excavated on each of these sites to the depth of consolidated
bedrock or 1.8 m, whichever was less. Five-kg soil samples
were taken from each horizon of each site. These were split,
with half used in a greenhouse bioassay and half used for
soil analysis. Soil analysis included texture (hand-textured
according to Thien 1979). Soils were taxonomically described in the field.
Seeds were collected from plants at four of the seven
sites. Tails were removed and the seeds submerged in a
0.525-percent sodium hypochlorite solution for 20 minutes
to control fungi. Sterile water-rinsed seeds were placed in
glass containers in wetted sterile vermiculite and refrigerated at 2 ∞C for 60 days, and then at 12 ∞C for 10 days. This
treatment resulted in 90-percent germination.
Autoclaved (121 ∞C at 0.1035 MPa for 20 minutes) pottery
shards were placed in the bottoms of plastic 30.5-cm-diameter pots to expedite drainage. Pots had been soaked overnight in a 0.525-percent sodium hypochlorite solution. Each
pot was half filled with autoclaved sand. Soil from each
horizon of the seven sites was each mixed 1:1 (weight to
weight) with autoclaved sand. The sand-soil mixture from
each specific horizon was used to layer 5 cm over the sand in
each pot. Over this an additional 5-cm layer of a 1:1 sand to
perlite mixture was added. Across the seven field sites, 23
86
different horizons were collected. Because two pots of diluted soil material were derived from each horizon, a total of
46 pots resulted. Two control pots were also constructed,
where a 1:1 sand to perlite mixture was substituted for the
1:1 sand to soil mixtures. Pots were thoroughly watered (to
near field capacity) and five germinated mountain mahogany seeds planted in the surface 2 cm of each container
and lightly covered with the sand to perlite mixture. All
plants were placed in a greenhouse in late April and misted
daily during the emergence and establishment period of the
plants. Pots were randomly placed, on 45-cm centers, in the
greenhouse and rerandomized monthly. Damping-off fungi
were problematic, and resulted in necessary replacement of
some seedlings. Pots were watered as needed. Temperatures
in the greenhouse averaged 20 to 22 ∞C during daylight
hours and 7 to 8 ∞C during night. Plants remained in this
environment for 10 months.
At the conclusion of this time, whole plants were removed
from the rooting media and observed for nodulation. Tops
and roots were separated and weighed individually.
Results ________________________
Soil Properties That Influence Actinorhizal
Development
For both mountain mahogany and antelope bitterbrush,
soil organic matter and extractable nitrate nitrogen are
strongly related to the capacity of these plants to host the
nitrogen fixing symbiont and consequent formation of
actinorhizae (see Peterson and others 1991; and especially
White and Williams 1985). Nodulated mountain mahogany
and antelope bitterbrush had, respectively, 0.61 (standard
deviation of 0.66) percent organic matter and 0.68 (0.59)
percent organic matter in root zones as compared to nonnodulated plants that were 2.94 (1.37) percent and 3.73
(2.24) percent. These differences were statistically significant between nodulated and non-nodulated populations
(within species) at 0.01 and 0.001, respectively, for mountain mahogany and antelope bitterbrush. Extractable nitrate nitrogen showed similar trends and significance levels—0.59 (0.50) ug/g and 0.92 (0.76) ug/g for nodulated
mountain mahogany and antelope bitterbrush, respectively; and 2.81 (1.73) and 3.50 (2.35), respectively, for nonnodulating plants. Differences were statistically significant at 0.01 levels for both comparisons within species.
Other soil parameters (pH, electrical conductivity, extractable P, and extractable K) were all significantly different
between nodulated and non-nodulated mountain mahogany, but were not significant for antelope bitterbrush.
These findings suggest that nodulation of these rosaceous
plants is controlled by edaphic factors such as organic
matter and nitrate levels. Given the high level of resources that a plant must invest in the acquisition of
nitrogen from atmospheric sources, it makes sense that
these plants would not allocate such resources if nitrogen
is available in mineral forms (for example, as nitrate) or
in organic forms (for example, from soil organic matter) at
levels above a basal level needed for survival. Artificial
fertilization of antelope bitterbrush (Tiedeman 1983) seedlings in a greenhouse study showed a positive response to
USDA Forest Service Proceedings RMRS-P-31. 2004
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
N, P, and S, but in the field, established mature plants did
not respond to fertilization with N, P, and S, although
associated understory plants did. Tiedeman’s work did not
examine the influence of fertilization on nodule formation.
Strategies for Plant Excavation
Techniques—Standard, nonstructured digging usually
results in holes that are roughly conical shaped with the flat
surface of the cone represented by the original surface of the
soil. This geometry is a result reflective of enthusiastic
excavation early and at the surface where digging is easy.
This activity wanes with time as the digger’s energy reserves
dissipate, and earthy substrates become more consolidated.
Our experiences have taught us to be somewhat and
sometimes opportunistic when seeking observations of the
belowground plant parts, especially of deeply rooted species
like mountain mahogany and antelope bitterbrush. Cuts
and trenches made by digging machinery can be excellent
places to observe roots of these plant species. Often these can
be accessed opportunistically when new roads or pipelines
are being constructed. It is possible and often desirable to
then work from the surface, progressing down the face of the
cut, excavating specific intervals of the substrate sequentially to the depth of the roots or to the depth of the trench
or cut. For mature mountain mahogany and antelope bitterbrush it is often convenient to stake off a distinct zone where
the shrub stem is roughly in the middle of the zone. We often
use a full square meter designation, where one side of the
square follows the edge of the cut or trench.
As excavation proceeds, it is usually necessary to remove
the aboveground portion of the plant. Some excavation can
be done with the plant in place, but this can be awkward. An
alternative is to divide the surface meter square into halfmeter or quarter-meter sectors and excavate those to depth.
In such cases, the aboveground plant parts can often be left
in place for part of the excavation.
We have generally found it expeditious to stake out the
area to be trenched and proceed as rapidly as possible with
the trenching. Certainly root observations can be made, but
it is also an option to ignore observations and extend the
trench to the length, depth, and width desirable. One should
be sure that the trench face, which will become the working
and access plane to the root system of the target plant, is
sufficiently far from the target so that root disturbance of
that individual is minimized. Having a level indicator for
such an excavation is of value to be sure the functional side
of the trench is vertical.
Before the careful excavation of the root system begins,
one should decide at what intervals soils should be removed.
It is often desirable to remove the soil by pedogenic horizon.
Sometimes this does not make sense if such horizons are
very thin, and at other times it may be desirable to subdivide
a particularly thick horizon. Intervals of 20 cm are often
convenient and seem to represent a scale that fits with the
size of these organisms.
In working with the first or top interval of the profile, one
can work from the surface of the exposed face back into
substratum to the extent of the predetermined boundaries of
the excavation. As this is done, photographs or maps of roots
and associated phenomena can be made. Nodules can be
observed and collected or tested (for example, acetylene or
USDA Forest Service Proceedings RMRS-P-31. 2004
Williams, Johnson, Munn, and Nieder
15
N incubations). It is best to completely remove a defined
volume of soil before moving to the next. It is usually best
also to sieve that material to catch missed roots and nodules
using a box sieve of usually 60 by 60 cm construction with
sieve openings of 1.25 cm. The substratum can be placed on
the sieve and agitated to pass through. Roots and nodules
will be caught on the sieve and recovered. The ease and
success of such an operation is very much connected with the
integrity of the substratum. Sand or gravelly material is
often easy to excavate and passes the sieve quickly and
efficiently. Decomposing granites can also work fairly well
and can be usually easily crushed to pass the sieve. High clay
content materials often require washing to remove roots as
do materials containing silts.
Usually as an excavation progresses to deeper depths,
coarse fragments tend to increase. This is especially true in
soils inhabited by mountain mahogany and antelope bitterbrush. Very coarse fragments are often very common especially in subsoils. Further, roots often do not abate when they
encounter consolidated rock. Roots may penetrate cracks,
and we have often come across nodules in cracks. Roots may
extend a meter or more into such materials. Nodules may be
found wherever roots go.
Examination of Soil Taxa Associated With Nodulated Shrub Communities—Soils identified along the
trench that intersected the mountain mahogany community
were Typic Haplustolls, loamy-skeletal, frigid under both
the grassland community adjacent to mountain mahogany
and under the mountain mahogany community. The soil
solum under both vegetation types consisted of noncalcareous
A and B horizons containing approximately 15 percent
course fragments overlying calcareous, skeletal C horizons.
The soils in this association were more highly developed
under the mountain mahogany than observed in other
locations where mountain mahogany was examined. The
major consideration at this site was a thin laminar petrocalcic
horizon at the surface of the skeletal layer in the soil under
the grassland community. The petrocalcic layer prevented
root penetration into the lower, high course fragment containing C horizon. Under the mountain mahogany stand
such a restrictive layer was not present, and roots penetrated to a considerable depth (to more than a meter). These
soils developed in limestone colluvium.
In many cases, mountain mahogany grows in very small
lenses of soil in cracks between rocks. From a taxonomic
standpoint these materials are essentially unconsolidated
rock. In the Rogers Canyon site it was noted that none of the
mountain mahogany whose roots were exposed in the trench
had nodules in the A horizon (the zone of highest organic
matter content); however, numerous plants whose root systems extended well below this horizon were nodulated in
subsoil horizons below the A horizon. These horizons were
very low in soil organic matter (less than 0.5 percent). In
other sites where plants were growing essentially in contact
with consolidated rock, nodules were seldom found; however, complete excavation of the root systems of these plants
was usually not possible.
In the antelope bitterbrush transect, the soils under the adjacent sagebrush grassland mixture were Typic Argiustolls,
fine-loamy, on upland sites grading into Cumulic
Haploustolls, fine-loamy, on sites (drainages) where there
87
Williams, Johnson, Munn, and Nieder
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
was transport of soil material onto the site. Bitterbrush
occupied Typic Torriorthents, loamy-skeletal, on the footslopes grading into Lithic Torriorthents, loamy-skeletal,
and eventually to rock on steeper sites.
Antelope bitterbrush, as is usually the case with mountain mahogany, occupied sites of shallower rocky soil than
associated vegetation (sagebrush and grasses). No nodules
were found where plant roots were in contact with high
organic matter soils (the A horizon). Nodules were found in
subsoils below the A horizon, although even these on this site
were sparse.
Nodulation Characteristics and Nitrogen
Fixation Activity
Actinorhizal Plants During Winter Months—Nodules encountered during the 1988 winter season had visual
signs (white tips, elastic and turgid) that suggested they
were active. However, all plants displayed necrotic nodules
(dark, shrunken, and brittle). Even the 1-year-old plant
had necrotic nodules (0.08 g) associated with active nodules
(0.05 g). The 5-year-old plants also had necrotic nodules (an
average of 0.13 g per plant) as well as active nodules
(average of 0.12 g dry weights 70 ∞C for 48 hours).
Ambient soil temperatures in the field and of the incubation chambers were just barely above the freezing point of
water (about 2 ∞C). Acetylene reduction assay indicated that
all nodules were inactive, even though some gave visual
appearances of activity. Ethylene produced by these nodules
was indistinguishable from background levels. Implications
here are that soil temperatures were too cold and plants too
inactive to provide photosynthate to accommodate nitrogen
fixation. However, an argument could be made from the
appearance of some nodules that they were poised to rapidly
begin nitrogen fixation once conditions or temperature and
photosynthate availability came more into functional ranges.
Biological Assay of Soil Materials for Nodulation
Capacity—Essentially the sites examined came from two
completely different parent materials. Sites 1, 2, and 3 were
on igneous and metamorphic rocks in the western part of the
county, whereas sites 4, 5, 6, and 7 were on sedimentary
rocks in the northern part of the county (table 1). The sites
were separated by approximately 40 m of elevation change
and 51 km of lateral change. The soils were Haplustolls and
Calciustolls on the upland, frigid sites (1, 2, and 3) and
Haplustolls on the lower elevation, mesic sites (4, 5, and 6).
Site 7 was a Torripsamment (Stevenson 2001).
Seedlings were susceptible to damping-off fungi. Although
soil material was dispersed between two pots for each
horizon tested, and a total of five seedlings were planted to
each pot, approximately 80 percent of the plants were lost,
although at least one plant survived in soil from each horizon
tested (table 1). Plants harvested at the end of the 10-month
period were observed for a variety of microbial manifestations on root systems, including the presence of nodules.
Roots were removed from the rooting media and stained
using an acid fuchsin technique (Williams 1979). Ecto-mycorrhizal and endomycorrhizal (arbuscular fungi) were observed
on all root systems. Infection density (mantled, swollen root
tips for ectomycorrhizae and arbuscules, vesicles and characteristic hyphae for endomycorrhizae) was variable, ranging
88
from light to very high in all horizons from sites 1 through 6.
Infection was unrelated to horizon or depth below the surface.
Mycorrhizal infections in roots exposed to the soil material
collected at site 7, where mountain mahogany was absent
from the flora, were present but the density was consistently
light.
Very small nodules were observed on bioassay roots exposed to subsoils from sites 1, 2, and 4. No other nodules were
observed.
After 10 months, most bioassay plants appeared stunted
and had very low fresh-weight biomasses (less than 0.5 g).
Large plants (greater than 1.0 g) were present, but were
always in soil material derived from A horizons or A/C
horizons. Some of these plants were more than 5.0 g fresh
weight. Nodules were never observed, however, on any
plants growing in A or A/C horizon materials. All nodules
were on plants growing in subsoil material derived from C
horizons. However, these nodulated plants were all small
plants.
The data from table 1 suggests that there were factors
present in the A horizons and sometimes the A/C horizons of
most soils examined that allowed mountain mahogany to
develop substantial growth. It is likely that these factors
were higher soil fertility associated with higher soil organic
matter in these horizons. However, sometimes a soil horizon
tested would generate a large plant and a stunted plant (for
example, site 1, horizon A generated plants of 5.64 and 0.39 g;
see table 1). Genetic variability among plants could also
account for these differences. Plants on materials derived
from C horizons almost always were very small, although a
few plants exceeded 0.5 grams and one plant exceeded 1.0 g
in size. Table 1 shows these average differences; however, it
does not show deviations from averages, many of which were
as large as the mean.
Table 1 also suggests that whatever factor was the causative agent of nodule formation was found only in subsurface
horizons (C material), but was inconsistent in those horizons
or was incompatible with some test plants. It would be
consistent with established theory that plants that became
nodulated would be large and robust. Here we suspect that
nodulated plants had not had sufficient time to respond to
nodulation, or that the nodules were ineffective. Another
season of growth may have resolved this issue, but this
would have required that nodulated plants be identified as
nodulated, a process that can only be done destructively.
Discussion _____________________
The work reported herein was conducted over several
decades and focused on acquiring field data on nodulation of
two Rosaceous shrubs, mountain mahogany, and antelope
bitterbrush.
From field observations, it is hypothesized that soil properties influence nodule initiation on both plants. Available
nitrogen, here as nitrate, and soil organic matter determine
nodulation in both plants. Although critical levels of nitrogen
and soil organic matter were not identified, relatively high
levels of nitrate nitrogen (approximately 3 ug/g) and soil
organic matter (2 to 4 percent) inhibited nodulation, whereas
low levels of nitrate (less than 1 ug/g) and soil organic matter
(less than 1 percent) were conducive to nodule formation. It is
USDA Forest Service Proceedings RMRS-P-31. 2004
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
Williams, Johnson, Munn, and Nieder
Table 1—Sites, locations, soil descriptions, and results of bioassay for mountain mahogany soils from Laramie County, Wyoming. Soil material from each horizon was
planted with mountain mahogany seedlings that were examined after 10 months of growth. A sand and perlite mixture (8) was planted as a control (see text).
Site
1. On calcareous igneous
colluvium (41∞11’5” N.;
105∞10’48” W.; 2,128 m,
15 percent north slope).
Native vegetation:
Achnatherum hymenoides,
Yucca glauca,
Cercocarpus montanus
(nodulated extensively).
2. On mixed calcareous
alluvium (41∞11’5” N.; and
105∞10’48”W., 2,134 m,
3 percent southwest slope).
Native vegetation:
Agropyron spp. and
C. montanus (nodulated)
3. On red sandstone over
white sandstone (41∞11’5” N.;
105∞10’48” W., 2,164 m,
20 percent west slope).
Native vegetation:
C. montanus (no nodules).
Cr is hard, white sandstone.
4. On very fine, noncalcareous
sandstone (41∞26’44” N.,
104∞40’22” W., 1,740 m, 20
percent north slope). Native
vegetation: Yucca glauca,
Eriogonium spp., Phlox spp.,
and nodulated C. montanus.
5. On semicalcareous
sandstone (41o26’44” N.,
04o40’22” W., 1,749 m, 30
percent west slope). Native
vegetation: Yucca glauca,
Phlox spp., and nodulated
C. montanus. Cr is semicalcareous sandstone .
6. On white sandstone
(41∞26’44” N.; 104∞40’22” W.,
1,750 m, 15 percent north
slope). Native vegetation:
Bouteloua gracilis, Ribes spp.,
Gutierrezia spp., and
nodulated C. montanus. Cr is
hard, calcareous sandstone.
7. On slightly calcareous
sandstone (41∞26’44” N.;
104∞40’22” W., 1,747 m,
10 percent southeast slope).
Native vegetation: mainly
Bouteloua gracilis and
Yucca glauca. No C.
montanus. Cr is hard,
calcareous sandstone.
8. Sand; perlite mixture.
Horizon
Depth
Surviving
plants
A
- - - cm - - 0 to 8
2
None
Nodulationa
Mycorrhizaeb
Variable
Average plant
fresh weightc
-----g----3.01
AC
8 to 30
2
None
From light
.21
C1
30 to 61
2
Yes
to very
.08
C2
61 to 91
3
Yes
high
.84
A
0 to 20
2
None
Bw
20 to 46
1
None
From light
C1
46 to 102
1
Yes
to very
.06
C2
102 to 152
2
None
high
.40
A
0 to 8
2
None
AC
8 to 30
3
None
From light
.08
C
30 to 56
2
None
to very
.05
Cr
56+
A
0 to 15
2
None
AC
15 to 51
2
None
From light
.25
C1
51 to 122
1
Yes
to very
.41
C2
122 to 183
1
None
high
.31
A
0 to 8
2
None
Variable
Variable
4.56
1.21
1.59
high
Variable
Variable
.73
.28
From
AC
8 to 28
2
None
light to
.42
very high
A
0 to 8
2
None
Variable
.38
AC
8 to 66
1
None
From
.24
C
66 to 102
1
None
Cr
102+
A
0 to 15
2
None
AC
25 to 30
2
None
Infection
C
30 to 122
1
None
only
Cr
122+
2
None
Control
light to
.39
very high
Light
No infection
.39
1.37
.07
.06
a
None, no nodules. Yes, nodules on at least one plant.
Ectomycorrhizae and endomycorrhizae were observed. Data not shown.
a
Fresh weight of whole plants.
b
USDA Forest Service Proceedings RMRS-P-31. 2004
89
Williams, Johnson, Munn, and Nieder
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
quite likely that the inhibitory effects of soil organic matter
and nitrate are closely related. Formation of nodules and
supplying them with photosynthate require considerable
energy inputs from the plant. It is likely that when sufficient
available nitrogen is present that feedback mechanisms in
the plant block nodule formation. The presence of soil organic
matter, which can be viewed as a form of potential labile
nitrogen, may also block nodule formation. It has also been
demonstrated that phenolic compounds, which are common
components of soil organic matter, often inhibit activity of
Frankia, the microbial causative agents of nodulation
(Perradin and others 1983).
Excavation of these nodulated Rosaceous plants in the
field needs to be done with care and attention to detail.
Strategies for excavation should involve allowing sufficient time and input of energy into the excavation process,
and excavation of mature plants may take even weeks to
accomplish.
Often recording accurate morphological data of roots and
nodules in the field can be done with some accuracy even if
the plant is being destroyed. Certainly the value of such data
is less with time as roots die, turgidity is lessened, and colors
are muted. However, to claim that a plant is responding
normally under such extreme conditions of stress is unjustified. Certainly, physiological data such as nitrogen fixation
activity, root uptake function, or photosynthesis cannot be
assumed to occur at normal rates while the plant is being
stressed during excavation. However, what roots do in soil
and what they do in unconsolidated or consolidated rock are
likely different. Nodules often and perhaps most likely on
some plants (for example, those of this report) occur very
deep below the soil surface. Mycorrhizal associations may
also occur very deeply in substrates. This poses a dilemma of
how to observe belowground biological phenomena in field
settings so that physical, chemical, and biological characteristics can be taken that are undistorted by the stress imposed on the biological system. Conversely, if a plant is
minimally sampled to minimize or even eliminate stress on
that plant, then there is uncertainty that the physical,
chemical, or biological observations being made are accurate.
Nodules observed and acetylene reduction made of bitterbrush during the winter months indicated that although
nodules gave visual appearance of activity, physiologically
they were not active in nitrogen fixation. Given the time of
year (winter), cold soil temperatures, and absence of photosynthetic organs on the plant tops (leaves), it is not surprising that the physiological function of the nodules was nil.
However, given that excavation of the plant was not a
normal condition of these plants, did excavation-induced
stress mask the physiological phenomena? Given the low
temperatures and shortened day length, it seems unlikely
that physiological response was being masked by plant
stress. A case might be made that plant physiological response may be more unfettered by excavation-induced stress
during these winter months than during any other time of
the year.
Evidence from the study where mountain mahogany was
grown in soil materials derived from pedogenic horizons
complements the work done and reported above, and suggests that soil properties (nitrate and soil organic matter)
inhibit nodulation. The work corroborates the observation
90
that mountain mahogany does not nodulate when in the
presence of soil materials having high organic matter. However, it provides a little evidence that active factor(s) that
cause nodulation reside within subsurface horizons very low
in soil organic matter. Members of the genus Frankia are
very slow growing (Callaham and others 1978; Stowers
1987). The fact that they would exist deep in soils, under low
carbon regimes and in zones of low microbial activity and
competition, is consistent with what we know about their
growth habit in controlled cultures. However, the results of
this part of the study really suggest only a hypothesis, not
currently adequately challenged: Frankia, responsible as
the microbial agent in nodulation of Rosaceous shrubs, are
active only in subsoils low in carbon and nitrogen.
A principle that has emerged from this particular work is
that there is a question that physiological measurements
represent normal plant response when the plant is being
stressed by the sampling process. Simultaneously, there is a
question that an adequate sample has been collected, especially of root parameters, if the stress due to sampling is
minimized or eliminated. Paradoxically, accurate measurements belowground that are reflective of a normal plant,
cannot be made because the plant must be stressed to make
such measurements. If the relationship between carbon
fixation and nitrogen fixation can be determined, then
monitoring carbon fixation during the excavation of the
plant can, at least theoretically, provide a means to correct
nitrogen fixation levels.
References _____________________
Allen, M. F. 1991. The ecology of mycorrhizae. Cambridge Studies
in Ecology (series). Cambridge University Press. 184 p.
Allison, L. E. 1965. Organic carbon. In: Black, C. A., ed. Methods of
Soil analysis. Agronomy 9: 1367–1378.
Bower, C. A.; Wilcox, L. V. 1965. Soluble salts. In: Black, C. A., ed.
Methods of soil analysis. Agronomy. 9: 933–951.
Callaham, D.; Tredici, P. Del; Torrey, J. G. 1978. Isolation and
cultivation in vitro of the actinomycete causing root nodulation in
Comptonia. Science. 199: 899–902.
Dalton, D. A.; Zobel, D. B. 1977. Ecological aspects of nitrogen
fixation by Purshia tridentata. Plant and Soil. 48: 57–80.
Joukowsky, M. 1980. A complete manual of field archaeology, tools
and techniques of field work for archaeologists. Prentice-Hall,
Inc. 630 p.
Kummerow, J.; Alexander, J. V.; Neel, J. W.; Fishbeck, K. 1978.
Symbiotic nitrogen fixation in Ceanothus roots. American Journal of Botany. 65: 63–69.
Marx, D. H. 1972. Ectomycorrhizae as biological deterrents to
pathogenic root infections. Annual Review of Phytopathology. 10:
429–454.
Oien, A.; Olsen, A. R. 1969. Nitrate determination in soil extracts
with the nitrate electrode. Analyst. 94: 888–884.
Perradin, Y.; Mottet, M. J.; Lalonde, M. 1983. Influence of phenolics
on in vitro growth of Frankia strains. Canadian Journal of
Botany. 61: 2807–2814.
Peterson, G. A.; Williams, S. E.; Moser, L. E. 1991. Chapter 18.
Inorganic fertilizer use and its effects on semiarid and arid region
soils. In: Skujins, J., ed. Semiarid lands and deserts: soil resource
and reclamation. Marcel Dekker: 543–580.
Reckner, R. 1998. Soil survey of Albany County area, Wyoming. U.S.
Department of Agriculture, Natural Resource Conservation Service, Forest Service, and Bureau of Land Management; and The
University of Wyoming Agricultural Experiment Station. 540 p.
plus maps.
Roskams, S. 2001. Excavation. Cambridge manuals in archaeology.
Cambridge University Press. 311 p.
USDA Forest Service Proceedings RMRS-P-31. 2004
Edaphic Characteristics of Nitrogen Fixing Nodulation (Actinorhizae) by Cercocarpus montanus ...
Soil Survey Staff. 1999. Soil taxonomy: a basic system of soil
classification for making and interpreting soil surveys, 2d ed.
Agric. Handb. 436. Washington, DC: U.S. Government Printing
Office.
Stahl, P. D.; Schuman, G. E.; Frost S. M.; Williams, S. E. 1998.
Arbuscular mycorrhizae and water stress tolerance of Wyoming
big sagebrush seedlings. Soil Science Society of America Journal.
62: 1309–1313.
Stevenson, A. 2001. Soil survey of Laramie County, Western Part.
U.S. Department of Agriculture, Natural Resource Conservation
Service; and The University of Wyoming Agricultural Experiment Station. 355 p. plus maps.
Stowers, M. D. 1987. Chapter 2. Collection, isolation, cultivation,
and maintenance of Frankia. In: Elkans, G. H., ed. Symbiotic
nitrogen fixation technology. Marcel Dekker, Inc.: 29–53.
Tiedemann, A. R. 1983. Response of bitterbrush and associated
plant species to broadcast nitrogen, phosphorus, and sulfur
fertilization. In: Tiedemann and Johnson, eds. Proceedings—
research and management of bitterbrush and cliffrose in Western
North America. Gen. Tech. Rep. INT-152. Ogden, UT: U.S.
USDA Forest Service Proceedings RMRS-P-31. 2004
Williams, Johnson, Munn, and Nieder
Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station: 240–254.
Thien, S. J. 1979. A flow diagram for teaching texture-by-feel
analysis. Journal of Agronomy. 8: 54–55.
Upchurch, R. G. 1987. Chapter 12. Estimation of bacterial nitrogen
fixation using acetylene reduction. In: Elkan, Gerald H., ed.
Symbiotic nitrogen fixation technology. Marcel Dekker, Inc.:
289–305.
Wantanabe, F. S.; Olsen, R. S. 1965. Test of an abscorbic acid
method for determining phosphorus in water and NaHCO3 extracts from soil. Proceedings: Soil Science Society of America. 29:
677–678.
White, J. A.; Williams, S. E. 1985. Symbiotic N2-Fixation by Frankia:
Actinorrhizae establishment on disturbed lands. Proceedings.
Denver, CO: American Society of Surface Mining and Reclamation, National Conference: 289–297.
Williams, S. E. 1979. Vesicular-arbsucular mycorrhizae associated
with actinomycete-nodulated shrubs, Cercarpus montanus Raf.
and Purshia tridentata (Pursh) DC. Botanical Gazette. 140(s):
S115–S119.
91
Germination
and Seedling
Establishment
Ipomopsis spicata ssp. robruthii
93
Illustration on the reverse side is by Walt Fertig. In: Wyoming Natural Diversity Database. 2000. Available at: http://www.uwyo.edu/wyndd (Accessed
12/18/03).
Cultural Methods for Enhancing
Wyoming Big Sagebrush Seed
Production
D. T. Booth
Y. Bai
E. E. Roos
Abstract: Demand for sagebrush seed is expanding, but wildland
harvesting is impaired by several factors including loss of stands to
weed invasions and fire. To increase the potential for seed production, sagebrush seed harvests were compared among cultural treatments for stands on a coal mine near Glenrock, WY. Sagebrush in
the study were mature plants on unmined rangeland (age unknown)
and 5- to 20-year-old plants on reclaimed land. Plants on mined land
protected by 1- by 1- by 1-m fence of 2.5-cm wire net had a 3-year
average seed yield of 20.7 g per plant compared to 5.2, 1.4, and 0.9
g per plant from plants on unmined-fenced, mined-unfenced, and
unmined-unfenced rangelands. Fencing prevented browsing by
herbivores and had a beneficial influence on plant microclimate.
Seed weight was greater where mulch or other treatments mitigated environmental stress. We conclude that the quantity and
quality of Wyoming big sagebrush seed harvest can be increased for
existing stands, particularly those on mined land, by protecting the
plants from large-animal herbivory and by using methods such as
windbreaks and fabric mulch to mitigate the effects of drought.
Introduction ____________________
Wyoming big sagebrush (Artemisia tridentata Nutt. ssp.
wyomingensis [Beetle & Young]) is used for ecological restoration of areas disturbed by fire, mining, cropping and other
situations requiring artificial revegetation with native
shrubs. That use has made Wyoming big sagebrush seed a
regionally important specialty crop with a history of sustained demand over more than a decade (R. Dunne, CEO,
Wind River Seed, personal communication; J. Johansen,
Natural Resource Specialist, BLM Regional Seed Warehouse - Boise District, personal communication). While the
demand for seed has been steadily increasing, native stands
suitable for seed harvesting have been steadily decreasing
due to fire and invasion by weeds. The situation suggests a
need to manage selected sagebrush stands, such as those on
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
D. T. Booth is a Rangeland Scientist, USDA-ARS, High Plains Grasslands
Research Station, 8408 Hildreth Road, Cheyenne, WY 82009, U.S.A., e-mail:
tbooth@lamar.colostate.edu. Y. Bai is an Assistant Professor, Department of
Plant Sciences, University of Saskatchewan, 51 Campus Drive, Saskatoon,
Saskatchewan S7N 5A8, Canada. E. E. Roos is Assistant Area Director,
USDA-ARS, Northern Plains Area, 1201 Oakridge Dr., Fort Collins, CO,
80525, U.S.A., formerly Research Leader and Plant Physiologist, National
Seed Storage Laboratory.
USDA Forest Service Proceedings RMRS-P-31. 2004
reclaimed mined lands, for improved seed production. Here
we review work conducted to access the quality and quantity
of seed that could be harvested from existing stands of
sagebrush at the Dave Johnston Coal Mine near Glenrock,
WY (Booth and others 2003).
Study Site Description ___________
The Dave Johnston Coal Mine is 40 km east of Casper, WY,
at an elevation of 1,646 m. The mean annual precipitation is
328 mm and the mean annual temperature is 8.8 ∞C (Martner
1986; Owenby and Ezell 1992) with an average 123 frost-free
days per year. Soil parent material is Cretaceous clay shale
(Young and Singleton 1977). Five sites were located on the
mine where mined-land sagebrush stands were in close
proximity to stands of sagebrush on unmined rangeland.
These sites ranged in age from 5 to more than 20 years at the
start of the study (table 1).
Experimental Design and
Treatments _____________________
The experimental design was a split plot, 2 by 2 factorial
replicated among the five sites with whole plots being mined
and unmined land at each site and subplots being cultural
treatments. Cultural treatments were assigned to single
sagebrush plants selected for uniformity and are described
as follows: (1) understory plants were cut to ground level,
2
and a 1-m piece of fabric mulch (Appleton and others 1990)
was installed around the base of the plant; (2) a windbreak
of fabric mulch was erected on the north and west side of the
plant, (3) both mulch and a windbreak were installed, and (4)
the selected plant was not treated (neither windbreak nor
mulch). Three subsamples (individual plants) were used for
each treatment combination. Treatments were installed in
July 1995.
2
A 1-m by 1-m-high fence of 2.54-cm wire net was placed
around plants selected for treatment. Fenced plants were
paired with a similar-size, unfenced, untreated adjacent
plant (table 1). Our test included an evaluation of any effects
of the wire net because we needed to protect experimental
plants from browsing wildlife, and we needed to measure the
effect of our protective measures.
Soil moisture was measured using gypsum blocks buried
to 10 cm below the soil surface and beneath the canopy of
treated plants. The blocks were read midmonth, May through
95
Booth, Bai, and Roos
Cultural Methods for Enhancing Wyoming Big Sagebrush Seed Production
Table 1—Experimental variables used in sagebrush seed production study.
Variable (number and identification)
Sites (5; Fuel Island, Entry 50, 110 School, 60 Badger, 4 School)
Whole plots (2; mined and unmined)
Subplots (4 treatments x [3 subsamples + 3 paired plots] = 24 as indicated below)
Mulch plots with paired untreated and unfenced plots (3 + 3)
Windbreak plots with paired untreated and unfenced plots (3 + 3)
Mulch + windbreak plots with paired untreated and unfenced plots (3 + 3)
Control plots with paired untreated and unfenced plots (3 + 3)
September 1996 through 1998. The data were analyzed by
month across the 3 years of the study.
Seeds were harvested annually in late October or early
November 1996 through 1998. Data collected included number of seed stalks and bulk weight of seeds produced per
plant; and seed weight, moisture content at harvest, germination, and seedling vigor. Seed quality tests used 20 seeds
per plant or the maximum number of seeds available if less
than 20. The methods used for seed-quality testing are
described elsewhere (Bai and others 1997; Booth and Griffith
1994; Booth and others 2003). We used the “Mixed” procedure in SAS to test for differences among variables and the
“differences” option to compare least-square means (Littell
and others 1996; SAS 1996). Protective fencing was evaluated in separate analyses as a subplot effect (table 2).
Results ________________________
Seed Yield
To our surprise, more seed was harvested from mined land
than from adjacent unmined rangeland. There was no difference between mined and unmined land in the number of seed
stalks counted at seed harvest, but plants on mined land
yielded more seed in all 3 years of the study. The extent of the
harvest was a factor of protective fencing. Fence-protected
plants produced greater seed stalk numbers (table 3), and
greater seed yield per plant (bulk weight). Protected plants
on mined land yielded 3 to 6 times more seed than protected
plants on unmined land (fig. 1).
Compared to the effects of land type and fencing, mulch
and windbreak had a relatively minor effect on yield. Mulch
was a significant (P < 0.10) factor in only 1 year and then as
part of a three-way interaction (table 4). Windbreak interacted with other factors to have a significant influence (P <
0.10) on seed yield in 2 of the 3 years.
Seed Quality
Fenced plants produced heavier seeds in 1996 and 1997,
but not in 1998 (table 3). Average seed dry weight had some
influence of land x mulch in 1997 and mulch in 1998 (table 5).
Seed moisture at harvest was similar among cultural treatments, but fenced plants consistently had higher seed moisture than unfenced plants (table 3). This may indicate
residual food storage activity in the seeds, and may account
for the trend for fenced plants to produce heavier seeds.
We found no differences in germination or dormancy due
to treatments, fencing, or types of land. Mean germination
percentages ranged between 39 and 92 percent from
unmined land, and between 77 and 92 percent from mined
land. Percentage dormant seeds (those that did not germinate but remained firm throughout the test period) ranged
from 3.5 to 23.7 percent in 1996, 1.0 to 6.8 percent in 1997,
and 1.2 to 8.4 percent in 1998. (Note the tendency for a
correlation between increasing seed dormancy with lower
seed yield [compare dormancy range with fig. 1]. It would
be interesting to learn if this apparent correlation is real.)
Means for seedling axial length in our test of seed vigor
ranged from 23 to 34 mm. There were no differences due to
land or fencing in 1997 or 1998 (P > 0.57), and in 1996 there
were not enough seeds harvested to compare treatments.
Table 2—Example programs, with model statements, as submitted to SAS for fenced, and fenced-versusunfenced plot analysis.
Fenced
Proc mixed;
Class site land mulch windbreak replication;
Model bulk weight =
Land
Mulch
Land x mulch
Land x windbreak
Land x mulch x windbreak;
Random site; site x land
Ismeans land mulch x windbreak
Land x mulch x windbreak/difference; run
96
Fenced versus unfenced
Proc mixed;
Class site land fence replication;
Model bulk weight =
Fence
Land
Land x fence;
Random site; site x land
Ismeans fence land x fence/difference; run
USDA Forest Service Proceedings RMRS-P-31. 2004
Cultural Methods for Enhancing Wyoming Big Sagebrush Seed Production
Booth, Bai, and Roos
Table 3—Effect of protective fencing on means of sagebrush seed stalks per plant, weight of sagebrush seeds, and seed moisture
1996 to 1998. The observed significance level is given for each pair of means.
1996
Variable
Fenced
Number seed-stalks
OSLa (df = 48)
Seed weight (mg per seed)
OSLa (df = 48)
Seed moisture (percent)
OSLa (df = 48)
a
22
4
0.0708
P = 0.01
3.9
1.3
P < 0.01
a
25
1996
1997
1998
a
20
15
a
10
b
5
b
b
b
b
0
Mined
fenced
Unmined
fenced
b
b
Mined
unfenced
b
b
Unmined
unfenced
Figure 1—Mean bulk weight of sagebrush seeds
per plant by year as affected by type of land and
protective fencing. The OSL for the land x fence
interaction was P = 0.03, 0.03, and 0.05 in 1996,
1997, and 1998. Means within a year followed by
the same letter are not different at P < 0.05 as
indicated by differences in least square means
within the “Mixed” statistical procedure (SAS 1996).
Land status
Mulch
Bulk weight
seeds
Windbreak
Mined
Mined
Mined
Mined
Unmined
Unmined
Unmined
Unmined
yes
No
No
Yes
Yes
Yes
No
No
No
Yes
No
Yes
Yes
No
No
Yes
--g-25.5a
21.8ab
13.0b
12.7b
4.3b
3.9b
3.1b
2.4b
1997
Mined
Mined
Unmined
Unmined
–
–
–
–
No
Yes
Yes
No
29.9a
20.1b
8.6b
6.0b
1998
Mined
Unmined
–
–
–
–
1996
10
P < 0.01
0.2483
0.1422
P < 0.01
5.6
3.1
P < 0.01
Fenced
Unfenced
76
15
P < 0.01
0.2325
0.1916
P = 0.29
4.0
2.5
P = 0.02
Soil Moisture (Fenced Plants Only)
Soil moisture in May was influenced by site x land x mulch
(P = 0.05), land x mulch x windbreak (P < 0.01), year x mulch
(P < 0.05), year x land (P < 0.01, and year x site (P < 0.01). The
interactions with site and year were expected. The threeway interactions (table 6) reveal that mined-land soils had
soil moisture equal to or greater than that of unmined lands.
The greatest soil moisture occurred on mined land with
mulch and windbreak.
June soil moisture interactions were site x land (P = 0.02),
year x mulch (P = 0.04), and year x site (P < 0.01). The site x
land data again indicated that mined-land soil moisture was
equal to, or greater than that of unmined lands (table 7).
There were no differences in main effects or interactions for
July (P > 0.83), and year x site was the only significant
interaction in August and September (P < 0.01 for both
months).
Plant Mortality
Table 4—Mean bulk weight of sagebrush seeds per fenced plant by
year as affected by type of land, mulch, and windbreaka.
Year
1998
Unfenced
56
P < 0.01
0.1507
Fenced
Observed significance level.
30
Seed harvested (g)
1997
Unfenced
19.1 (P = 0.10)
4.4
a
Observed significance level: 1996 land x mulch x windbreak P = 0.06; 1997
land x windbreak P = 0.07; and 1998 land P = 0.10; df: land = 4, windbreak = 102,
mulch = 102, interactions = 104. Means within a year followed by the same letter
are not different at P < 0.05 as indicated by differences in least-square means
within the “Mixed” statistical procedure (SAS 1996).
USDA Forest Service Proceedings RMRS-P-31. 2004
There were three mined-land plants in the study that died
at 110 School (12.5 percent) and 11 that died at 60 Badger (46
percent). (Note the high soil moisture for mined land at 60
Badger [table 7].) All mortality during the 3 years occurred
on mined land; there was no mortality at any of the other
study sites.
Table 5—Seed quality: mean weight of sagebrush seeds by year as
affected by land type and mulcha.
Year
Land status
1996
Mined
Unmined
Unmined
Mined
Mined
Unmined
–
–
1997
1998
Mulch
Seed weight
–
–
Yes
Yes
No
No
Yes
No
- - -mg - - 0.21
0.11
0.33a
0.28ab
0.27ab
0.24b
0.28
0.24
OSLb
P = 0.22
P = 0.09
P = 0.09
a
Multiple means within a year followed by the same letter are not different at
P < 0.05 as indicated by differences in least-square means using the “Mixed”
statistical procedure (SAS 1996). Degrees of freedom: land = 4, mulch = 104,
windbreak = 104, interactions = 104.
b
OSL = Observed significance level.
97
Booth, Bai, and Roos
Cultural Methods for Enhancing Wyoming Big Sagebrush Seed Production
Table 6—Mean percentage soil moisture for May averaged over
5 years and as affected by land, mulch, and windbreak.
Windbreak
Yes
Yes
No
No
a
Mulch
Yes
No
OSL for mulch
Yes
No
OSL for mulch
Mined
Unmined
- - - - percent - - - 72.8
62.6
59.6
55.6
<0.01
<0.01
69.3
66.5
69.7
58.4
.87
<0.01
OSLa for land
<0.01
.09
.29
<0.01
Observed significance level.
Table 7—Mean percentage soil moisture for June averaged over 3 years
as affected by site and land.
Site
Fuel Island
Entry 50
110 School
60 Badger
4 School
a
Mined
Unmined
- - - - - - percent - - - - - - 20
16
36
27
31
37
70
45
35
21
OSLa for land
0.55
.16
.36
<0.01
.03
Observed significance level.
Discussion _____________________
Why Were Seed Yields Greater on Mined
Versus Unmined Land?
Seed yields from fenced plants on mined land were consistently greater than for other treatment combinations. This
may have been due to soil moisture, which was usually
greater under mined-land plants in May and June. Why
these soils contained more moisture is a matter for speculation. Conceivably plant density—either of shrubs or of all
plants—was lower on mined land. Plant density often affects
seed yield and differences in soil moisture may be only one
aspect of that affect (Henderson and others 2000; Holen and
others 2001; Lopez-Bellido and others 2000; USDA 1961).
Given that seed quality as measured by seed weight, seed
moisture, germination, and seedling vigor (axial length) was
not affected by land status, the greater seed yield for mined
land suggests seed producers might wish to direct cultural
efforts to mined-land stands.
Fenced and Unfenced Plants
It is clear that large herbivores reduced seed production by
eating seed stalks (table 4). Wagstaff and Welch (1991) also
reported that protected sagebrush plants produce significantly more seed stalks than plants not protected from
grazing animals. We recognize that protective fencing modified the environment in favor of the protected plants. However, the report of Gores (1995), and personal observations
of the effects of wildlife on mined-land shrub stands suggest
the bias was relatively insignificant and should not be used
98
to underrate wildlife herbivory as a factor limiting seed
production of mined-land sagebrush.
Environmental Modification by Fence,
Windbreak, and Mulch
We found that surrounding a sagebrush plant with wire
netting (fenced) resulted in a beneficial modification of its
environment as evident by greater seed weights from protected plants (table 4). This was not expected, especially
with the large (2.54-cm) net used. The fencing affect may
account for the minimal effect of mulch and windbreak on
seed production and quality. Mulch and windbreak did
influence seed production and quality, but these effects were
2
not consistent. The mulch, due to its small size (1 m ), did not
always result in greater soil moisture (table 7). A larger piece
of fabric would have shown a more consistent benefit to soil
moisture, shrub growth, and seed production (Booth, unpublished data).
Conclusions and Recommendations
Wyoming big sagebrush on reclaimed mined lands similar
to those of the Dave Johnston Coal Mine have the potential
to produce more seeds than plants from adjacent unmined
land. However, the mined-land seed-production advantage
will be realized only when the plants are protected from
browsing wildlife. Environmental modification—particularly the use of fabric mulch—can also improve seed production and some factors of seed quality, but these efforts are
viewed as secondary to that of excluding browsing animals.
Acknowledgments ______________
The work was supported in part by the Abandoned Coal
Mine Lands Research Program at the University of Wyoming. Support was administered by the Wyoming Department of Environmental Quality from funds returned to
Wyoming from the Office of Surface Mining of the U.S.
Department of the Interior. The authors thank C. Skilbred,
Sr., Environmental Scientist for Glenrock Coal Company for
cooperation in this study and L. W. Griffith for technical
assistance.
References _____________________
Appleton, B. L.; Derr, J. F.; Ross, B. B. 1990. The effect of various
landscape weed control measures on soil moisture and temperature, and tree root growth. Journal of Arboculture. 16: 264–268.
Bai, Y.; Booth, D. T.; Roos, E. E. 1997. Effect of seed moisture on
Wyoming big sagebrush seed quality. Journal of Range Management. 50: 419–422.
Booth, D. T.; Bai, Y. 1999. Effects of imbibition temperature on
seedling vigor: comparison among crops and rangeland shrubs.
Journal of Range Management. 52: 534–538.
Booth, D. T.; Bai, Y.; Roos, E. E. 2003. Wyoming big sagebrush seed
production from mined and unmined rangelands. Journal of
Range Management. 56: 542–546.
Booth, D. T.; Gores, J. K.; Olson, R. A.; Schuman, G. E. 1999. Shrub
densities on pre-1985 reclaimed mine lands in Wyoming. Restoration Ecology. 7: 24–32.
Booth, D. T.; Griffith, L. W. 1994. Measuring post-germination
growth. Journal of Range Management. 47: 503–504.
USDA Forest Service Proceedings RMRS-P-31. 2004
Cultural Methods for Enhancing Wyoming Big Sagebrush Seed Production
Gores, J. D. 1995. The evaluation of big sagebrush and fourwing
saltbush establishment on pre-1985 Wyoming reclaimed mine
sites. Laramie: University of Wyoming. 150 p. Thesis.
Henderson, T. L.; Johnson, B. L.; Schneiter, A. A. 2000. Row spacing,
plant population, and cultivar effects on grain amaranth in the
Northern Great Plains. Agronomy Journal. 92: 329–336.
Holen, D. L.; Bruckner, P. L.; Martin, J. M.; Carlson, G. R.; Wichman,
D. M.; Berg, J. E. 2001. Response of winter wheat to simulated
stand reduction. Agronomy Journal. 93: 364–370.
Littell, R. C.; Milliken, G. A.; Stroup, W. W.; Wolfinger, R. D. 1996.
SAS system for mixed models. Cary, NC: SAS Institute, Inc.
Lopez-Bellido, L.; Fuentes, M.; Castillo, J. E. 2000. Growth and yield
of white lupine under Mediterranean conditions. Agronomy Journal. 92: 200–205.
Lyford, M. E. 1995. Shrub establishment on drastically disturbed
lands. Laramie: University of Wyoming. 84 p. Thesis.
USDA Forest Service Proceedings RMRS-P-31. 2004
Booth, Bai, and Roos
Martner, B. E. 1986. Wyoming climate atlas. Lincoln: University of
Nebraska Press.
Owenby, J. R.; Ezell, D. S. 1992. Monthly station normals of
temperature, precipitation, and heating and cooling degree days,
1961–1990 for Wyoming. Asheville, NC: National Climatic Data
Center.
SAS. 1996. SAS/STAT software: changes and enhancements through
release 6.11. Cary, NC: SAS Institute, Inc.
USDA. 1961. Seeds—the yearbook of agriculture. Washington, DC:
U.S. Government Printing Office.
Wagstaff, F. J.; Welch, B. L. 1991. Seedstalk production of mountain
big sagebrush enhanced through short-term protection from
heavy browsing. Journal Range Management. 44: 72–74.
Young, J. F.; Singleton, P. C. 1977. Wyoming general soil map.
Wyoming Agricultural Experiment Station Bulletin. Research
Journal 117. Laramie: University of Wyoming.
99
Seeds of Success and the
Millennium Seed Bank Project
Ann DeBolt
Carol S. Spurrier
Abstract: The Bureau of Land Management (BLM) entered into an
agreement with the Royal Botanic Gardens, Kew (RBG, Kew), to
collect seeds from native plants in the Western United States. This
program, initiated in 2000, is called “Seeds of Success.” It is an
offshoot of the international “Millennium Seed Bank” project. The
goal of the Millennium Seed Bank project is to collect seed from10
percent of the world’s dryland flora and place it in long-term storage.
For the BLM, this was an opportunity to evaluate additional native
pant materials for restoration projects and to place seed in longterm storage for conservation. Collection targets included native
species: of known forage or browse value, widespread regional
endemics, wild relatives of cultivated or economically important
species, species with significance to Tribes, monotypic species,
relatives of rare species, relatives of nonnative invasive species,
species of known pollinator importance, and flagship species such as
State flowers and trees. RBG, Kew, and the BLM developed a seed
collecting protocol at the population level. In 2002, the first collecting season, BLM gathered seed from 375 taxa in 11 States. Seed is
currently in long-term storage at RBG, Kew, and the USDA National Seed Storage Laboratory in Fort Collins, CO.
Introduction ____________________
The Royal Botanic Gardens, Kew (RBG, Kew), has initiated a worldwide seed collection program called the “Millennium Seed Bank” project. It is a multiyear project funded by
the United Kingdom Millennium Trust with a goal of collecting and conserving at least 10 percent of the world’s flora
(approximately 24,000 species) by the year 2010. Fourteen
countries around the world are currently participating in
the project. The Bureau of Land Management (BLM) participates with the RBG, Kew, and the Plant Conservation
Alliance in an offshoot program named “Seeds of Success.”
The goal of Seeds of Success is multifold. Establishing a
high-quality, accurately identified, and well-documented
native seed collection at the USDA National Seed Storage
Laboratory (NSSL) in Fort Collins, CO, with a backup
collection at the Millennium Seed Bank, is a major objective.
Additional, equally important objectives are to identify
important native species for restoration of public lands, and
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Ann DeBolt is Botanist at the U.S. Forest Service, Rocky Mountain
Research Station, 316 East Myrtle, Boise, ID 83702, U.S.A.; e-mail:
a.debolt@fs.fed.us. Carol S. Spurrier is Botany Program Lead at BLM’s
Washington Office, 1849 C Street NW, Washington, DC 20240, U.S.A.
100
to obtain information on the quality, viability, and germination requirements of those species.
The BLM was an obvious choice for RBG, Kew, to contact
given the extensive arid regions the agency manages. The
BLM and RBG, Kew, participate in the Seeds of Success
program under the terms of a cooperative agreement signed
by both parties in May 2000. The BLM agreed to collect seeds
for the project and to grant access to the lands they manage
for collection; to grant prior informed consent to RBG, Kew,
for study and long-term storage of these seeds; and to send all
seeds and vouchers to Kew. RBG, Kew, agreed to clean,
process, test, and store all seed sent by the BLM; to return half
of each collection to the United States for long-term storage;
to provide the BLM with all testing results; to fund a fixedterm coordinator position in BLM to develop the collection
program; and to provide training and advice to BLM during
the project.
RBG, Kew, is interested in seed banking a range of species
for long-term conservation of genetic material in support of
the Convention on Biodiversity, and because their funding is
partly provided by their ability to seed bank at the species
level. Given that seeds from the majority of arid land plants
in the United States are expected to survive conventional
germplasm bank conditions (that is, drying to low moisture
content and freezing at –20 ∞C), land managers can rely on
conserved seed samples for use in plant conservation and
restoration programs. As BLM collects seeds for restoration
across the range of the species, they work with local seed
growers to ensure that private businesses are given the
opportunity to use and market native plant materials originating from the project.
As a multiple-use agency, the BLM has a need for native
plant materials for many different programs. One of the
goals set forth in the BLM’s mission statement is to restore
at-risk resources and to maintain or improve functioning
ecosystems. In 1999 and 2000, wildfires burned more than 3
million acres, creating a demand that far exceeded the
supply of native species. Fires often increase establishment
of exotic, invasive species, and vast acreages are in need of
restoration. Native seed is needed for BLM’s fuels reduction
program, for prescribed fire planning and implementation,
and for the Great Basin Restoration Initiative. Wildlife
enhancement programs, particularly those for threatened or
endangered species, require native seed. Supplying energy
is currently a priority activity for BLM, and native plant
materials are needed for mitigation in mining reclamation
and oil and gas exploration activities, and for nonenergyrelated rights-of-way restoration such as communication
lines and new roads. The newly designated national monuments have requirements for the use of native plant species
USDA Forest Service Proceedings RMRS-P-31. 2004
Seeds of Success and the Millennium Seed Bank Project
DeBolt and Spurrier
inside monument boundaries. Seeds of Success can help to
successfully build a native plant materials program across
the United States. It will also enhance the seed conservation
capacity currently available within the United States, principally at the USDA NSSL in Fort Collins, CO.
Methods _______________________
The collecting focus for Seeds of Success is on species
needed for ecological restoration of public lands, native plant
material development, and conservation of widespread species. Target species lists were developed using The Nature
Conservancy’s ecoregions and included native species of
known forage or browse value, widespread regional endemics,
wild relatives of cultivated or economically important species, species with significance to Tribes, monotypic species,
relatives of rare species, relatives of nonnative invasive
species, species of known pollinator importance, and wellknown species that the public can easily identify, such as
State flowers and trees.
According to Brown and Marshall (1995), at least one copy
of 95 percent of the alleles occurring in a population at
frequencies of greater than 0.05 can be achieved by sampling
from:
1. 30 randomly chosen individuals in a fully outbreeding
sexual species, or
2. 59 randomly chosen individuals in a self-fertilizing
species.
Because the reproductive biology of most target species has
not been studied, and the capture of rarer alleles would
require a markedly increased sample size, collectors are
advised to sample from in excess of 50 individuals within a
single population, and to look for populations with larger
numbers of plants. For Seeds of Success, it was determined
that an “ideal” collection would be from 100 to 500 individuals and contain 10,000 to 20,000 seeds. Each Western State
was assigned a target number of species to collect during
the first year. To help accomplish this, the Student Conservation Association (SCA) partnered with BLM and hired
teams of three to four interns to work with botanists in five
States, including California, Idaho, Oregon, Nevada, and
Utah (fig. 1). The BLM also entered into a cooperative
agreement with the Center for Plant Conservation to use
their seed collection expertise in reaching BLM’s goal of
collecting 2,000 species over the next 10 years.
Once target species were identified, but before seed could
be collected, potential sites were visited to assess the population size, collect a herbarium voucher for species verification (nomenclature follows Kartesz and Meacham 1999),
evaluate presence of insect damage, and estimate seed
production and maturation date. Seed collection protocols
are as follows:
1. Carefully examine a small, representative sample of
seeds using a cut test to estimate the frequency of empty or
damaged seeds, and to confirm that the majority of seeds are
mature. Repeat cut tests throughout the sampling to ensure
seed quality across the population.
2. Collect mature, dry seeds into either cloth or brown
paper bags. Large collections could be made using plastic
buckets, for later transfer.
USDA Forest Service Proceedings RMRS-P-31. 2004
Figure 1—Student Conservation Association interns
collecting seed in Idaho for the Bureau of Land
Management Seeds of Success program.
3. If seeds can be liberated from their fruits quickly and
easily, do so and record this on the data form. Cleaning
should generally be left to RBG, Kew, staff.
4. Collect fleshy fruits into plastic bags and allow to
aerate. Contact RBG, Kew, right away for specific advice on
ripening and cleaning, as these fruits decompose rapidly and
poor storage can lead to mold-infested seed.
5. Sample equally and randomly across the extent of the
population, maintaining a record of the number of individuals sampled.
6. Collect no more than 20 percent of the viable seed
available on the day of collection.
7. Collect from populations that allow removal of 10,000
to 20,000 viable seeds.
8. For each collection, estimate and record the viable seed
production per fruit, per individual, and per population.
After collecting, seeds were kept in a cool, dry place prior
to shipping to the seed bank. Voucher specimens and data
forms were shipped with the seed to RBG, Kew, typically
within a few days. Data recorded for each population included species name, collector name(s), location (narrative
description plus latitude/longitude), elevation, habitat, landform, parent material, slope, soil texture, number of plants
sampled, an estimate of the total population size and area,
and the number of duplicate voucher specimens made. The
location of each population was recorded with a Global
Positioning System (GPS) unit, and digital photos of each
species and its seed were taken. After cleaning and testing,
half of each sample shipped to the RBG, Kew, seed bank will
return to the United States for native plant materials
development or long-term storage.
Results ________________________
Seed from 375 native taxa were collected in 2002, the
first year of implementation for Seeds of Success (table 1).
Taxa collected in more than one state are listed as such.
Student Conservation Association intern teams assisted in
five of the 11 States where collections were made, and
101
DeBolt and Spurrier
Seeds of Success and the Millennium Seed Bank Project
Table 1—Taxa and State of origin for 2002 native seed collections for the Bureau of Land Management’s Seeds of Success program. Seed is currently
in long-term storage at RBG, Kew, as part of the international Millennium Seed Bank project, or in the Bureu of Land Management’s native
plant materials development program.
Family
Aceraceae
Anacardiaceae
Anacardiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Apiaceae
Asclepiadaceae
Asclepiadaceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Genus
Acer
Rhus
Rhus
Angelica
Cnidium
Heracleum
Lomatium
Lomatium
Lomatium
Lomatium
Lomatium
Lomatium
Osmorhiza
Perideridia
Asclepias
Asclepias
Achillea
Achillea
Agoseris
Agoseris
Ambrosia
Antennaria
Arnica
Arnica
Arnica
Arnica
Artemisia
Artemisia
Artemisia
Artemisia
Aster
Baccharis
Balsamorhiza
Balsamorhiza
Balsamorhiza
Balsamorhiza
Bebbia
Blepharipappus
Brickellia
Brickellia
Chaenactis
Chrysothamnus
Cirsium
Cirsium
Crepis
Encelia
Erigeron
Erigeron
Erigeron
Erigeron
Erigeron
Erigeron
Eriophyllum
Eriophyllum
Eriophyllum
Eurybia
Eurybia
Grindelia
Helianthella
Heterotheca
Hieracium
Species
negundo
trilobata
trilobata v. trilobata
kingii
cnidiifolium
lanatum
dissectum
grayi
macrocarpum
nudicaule
nuttallii
triternatum
berteroi
bolanderi
cryptoceras
speciosa
millefolium
millefolium v. occidentalis
glauca v. glauca
heterophylla
eriocentra
monocephala
chamissonis ssp. foliosa
cordifolia
lessingii ssp. lessengii
longifolia
tridentata v. tridentata
arctica ssp. arctica
suksdorfii
tilesii ssp.elatior
pauciflorus
pilularis
hookeri
macrophylla
sagittata
serrata
juncea
scaber
incana
oblongifolia
douglasii
humilis
cymosum
mohavensis
acuminata
virginensis
bloomeri
disparipilus
glaucus
peregrinus
pumilus
speciosus
lanatum v. grandiflorum
lanatum v. integrifolium
staechadifolium
radulina
sibirica
squarrosa
uniflora
villosa
triste
State
UT
UT
CO
NV
AK
CA
ID
ID
CA
CA
CO
CA
OR
OR
OR
ID
CA
OR
OR
OR
NV
AK
CA
OR, UT
AK
UT
UT
AK
CA
AK
NV
OR
ID
UT
CA, ID
OR
NV
CA, OR
NV
NV
ID
OR
CA
NV
ID
NV
ID
ID
CA
AK
ID
UT
CA
CA
CA
OR
AK
UT
UT
CO, UT
AK
(con.)
102
USDA Forest Service Proceedings RMRS-P-31. 2004
Seeds of Success and the Millennium Seed Bank Project
DeBolt and Spurrier
Table 1—(Con.)
Family
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Berberidaceae
Betulaceae
Bignoniaceae
Boraginaceae
Boraginaceae
Boraginaceae
Boraginaceae
Brassicaceae
Brassicaceae
Brassicaceae
Brassicaceae
Cactaceae
Campanulaceae
Campanulaceae
Campanulaceae
Caprifoliaceae
Caprifoliaceae
Caprifoliaceae
Chenopodiaceae
Chenopodiaceae
Chenopodiaceae
Chenopodiaceae
Chenopodiaceae
Cornaceae
Cornaceae
Crassulaceae
Cupressaceae
Cupressaceae
Cupressaceae
Cucurbitaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Genus
Hymenoxys
Isocoma
Iva
Microseris
Petasites
Peucephyllum
Psilocarphus
Rudbeckia
Rudbeckia
Senecio
Senecio
Solidago
Solidago
Solidago
Stephanomeria
Stephanomeria
Symphyotrichum
Taraxacum
Tetradymia
Tetradymia
Townsendia
Tripleurospermum
Wyethia
Wyethia
Wyethia
Xylorhiza
Mahonia
Betula
Chilopsis
Cryptantha
Heliotropium
Lithospermum
Mertensia
Draba
Lepidium
Polyctenium
Stanleya
Opuntia
Downingia
Downingia
Downingia
Lonicera
Lonicera
Symphoricarpos
Atriplex
Atriplex
Grayia
Nitrophila
Suaeda
Cornus
Cornus
Sedum
Juniperus
Juniperus
Juniperus
Cucurbita
Carex
Carex
Carex
Carex
Carex
Carex
Carex
Species
cooperi
drummondii
acerosa
troximoides
hyperboreus
schottii
oregonus
laciniata v. ampla
occidentalis
atratus
spartioides v. spartioides
canadensis
multiradiata v. multiradiata
spathulata v. spathulata
exigua
pauciflora
spathulatum
species not yet determined
canescens
spinosa
florifer
maritimum ssp. phaeocephalum
amplexicaulis
mollis
scabra
tortifolia
aquifolium ssp. aquifolium
occidentalis
linearis
circumscissa
curassavicum
ruderale
ciliata
verna
dictyotum
fremontii
pinnata
whipplei
bacigalupii
bicornuta
insignis
hispidula
involucrata
oreophilus
canescens
confertifolia
spinosa
occidentalis
torreyana
nuttallii
sericea
spathulifolium
communis
osteosperma
occidentalis
palmata
alma
aurea
interior
microptera
nebrascensis
pellita
praegracilis
State
NV
UT
UT
ID, OR
AK
CA, NV
CA
CO
UT
CO
NV
UT
AK
CA
OR
NV
ID
AK
ID
ID
ID
AK
ID
CA
UT
NV
OR
UT
NV
ID
NV, OR
ID
UT
ID
CA
OR
NV
NV
CA
OR
CA
OR
UT
UT
NV
NV
CA
NV
NV
OR
CO
OR
NV
NV
ID
NV
NV
NV
UT
NV
UT
UT
NV, UT
(con.)
USDA Forest Service Proceedings RMRS-P-31. 2004
103
DeBolt and Spurrier
Seeds of Success and the Millennium Seed Bank Project
Table 1—(Con.)
Family
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Elaeagnaceae
Ephedraceae
Ericaceae
Ericaceae
Ericaceae
Ericaceae
Ericaceae
Ericaceae
Ericaceae
Ericaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Fabaceae
Garryaceae
Gentianaceae
Grossulariaceae
Grossulariaceae
Grossulariaceae
Grossulariaceae
Hydrangeaceae
Hydrophyllaceae
Hydrophyllaceae
Hydrophyllaceae
Iridaceae
Iridaceae
Iridaceae
Iridaceae
Iridaceae
Juncaceae
Juncaceae
Juncaceae
Genus
Cladium
Eleocharis
Eleocharis
Eleocharis
Fimbristylis
Schoenoplectus
Schoenoplectus
Schoenoplectus
Schoenoplectus
Schoenoplectus
Schoenus
Scirpus
Trichophorum
Shepherdia
Ephedra
Arctostaphylos
Arctostaphylos
Arctostaphylos
Harrimanella
Phyllodoce
Rhododendron
Vaccinium
Vaccinium
Acacia
Astragalus
Astragalus
Astragalus
Astragalus
Astragalus
Astragalus
Astragalus
Astragalus
Astragalus
Cercis
Glycyrrhiza
Hedysarum
Lathyrus
Lupinus
Lupinus
Prosopis
Prosopis
Prosopis
Prosopis
Sophora
Thermopsis
Garrya
Gentiana
Ribes
Ribes
Ribes
Ribes
Whipplea
Eriodictyon
Phacelia
Phacelia
Iris
Iris
Sisyrinchium
Sisyrinchium
Sisyrinchium
Juncus
Juncus
Juncus
Species
californicum
palustris
parishii
rostellata
thermalis
acutus
americanus
maritimus
pungens v. longispicatus
tabernaemontani
nigricans
microcarpus
alpinum
argentea
californica
canescens ssp. canescens
columbiana
patula
stelleriana
glanduliflora
camtschaticum ssp. spathulata
ovatum
parvifolium
greggii
alpinus
filipes
lentiginosus
lentiginosus v . floribundis
obscurus
preussii ssp. preussii
purshii ssp. purshii
utahensis
whitneyi
orbiculata
lepidota
alpinum
littoralis
arbustus
argenteus
gladulosa
gladulosa v. torreyana
pubescens
velutina
stenophylla
californica v. argentea
elliptica
glauca
cereum v. cereum
cruentum v. cruentum
sanguineum v. glutinosum
sanguineum v. sanguineum
modesta
angustifolium
capitata
humilis ssp. humilis
douglasiana
missouriensis
demissum
halophilum
species not yet deterrmined
alpinus
balticus
breweri
State
NV
ID
NV
UT
NV
UT
NV, UT
NV
UT
NV
NV
OR
AK
UT
CA
OR
OR
OR
AK
AK
AK
CA, OR
OR
NV
AK
ID
CA
CA
ID
UT
CA
UT
CA
NV
UT
AK
CA
CA
CA, UT
NV
NV
NV
AZ
UT
CA
OR
AK
CA
OR
CA
OR
OR
NV
OR
CA
CA
CA, CO
UT
NV
NV
UT
ID, NV, UT
CA
(con.)
104
USDA Forest Service Proceedings RMRS-P-31. 2004
Seeds of Success and the Millennium Seed Bank Project
DeBolt and Spurrier
Table 1—(Con.)
Family
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaceae
Juncaginaceae
Juncaginaceae
Lamiaceae
Lamiaceae
Lamiaceae
Lamiaceae
Lamiaceae
Lamiaceae
Lamiaceae
Lamiaceae
Liliaceae
Liliaceae
Liliaceae
Liliaceae
Liliaceae
Liliaceae
Liliaceae
Lillaceae
Lillaceae
Lillaceae
Loasaceae
Loasaceae
Loasaceae
Malvaceae
Malvaceae
Malvaceae
Myricaceae
Nyctaginaceae
Oleaceae
Onagraceae
Onagraceae
Onagraceae
Orchidaceae
Orchidaceae
Papaveraceae
Plumbaginaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Genus
Juncus
Juncus
Juncus
Juncus
Juncus
Juncus
Juncus
Juncus
Juncus
Luzula
Triglochin
Triglochin
Agastache
Hedeoma
Lepechinia
Mentha
Monardella
Salvia
Stachys
Stachys
Allium
Calochortus
Camassia
Camassia
Lilium
Lilium
Maianthemum
Veratrum
Zigadenus
Zigadenus
Eucnide
Mentzelia
Mentzelia
Sphaeralcea
Sphaeralcea
Sphaeralcea
Morella
Abronia
Fraxinus
Clarkia
Epilobium
Epilobium
Epipactis
Platanthera
Argemone
Armeria
Achnatherum
Achnatherum
Achnatherum
Andropogon
Aristida
Bromus
Bromus
Bromus
Calamagrostis
Danthonia
Distichlis
Elymus
Festuca
Melica
Muhlenbergia
Muhlenbergia
Panicum
Species
castaneus
cooperi
drummondii
effusus v. brunneus
longistylis
mertensianus
nodosus
saximontanus
torreyi
parviflora
concinna
maritimum
urticifolia
nana
calycina
arvensis
odoratissima
sonomensis
ajugoides v. rigida
palustris
acuminatum
macrocarpus
leichtlinii
quamash
columbianum
pardalinum ssp. pardalinum
stellatum
viride
elegans
venenosus v. venenosus
urens
albicaulis
lacinata
grossulariifolia
munroana
parviflora
californica
latifolia
velutina
gracilis
ciliatum
species not yet determined
gigantea
dilatata
hispida
maritima v. californica
hymenoides
nelsonii
thurberianum
glomeratus
purpurea
carinatus
marginatus
vulgaris
nutkaensis
californica
spicata
spicata
californica
californica
asperifolia
rigens
virgatum
State
AK
NV
AK
OR
UT
AK
NV, UT
UT
UT
AK
NV
UT
CA, UT
NV
CA
UT
CA, UT
CA
CA
UT
ID, OR, UT
OR
OR
OR
OR
OR
NV
AK
AK
CA
NV
ID
NV
ID
UT
UT
OR
OR
NV
CA
UT
AK
NV
AK
CO
CA
UT
CA
CA, ID
NV
UT
OR
CA
OR
CA
CA, OR
NV
CA
CA, OR
CA
NV, UT
NV
NV
(con.)
USDA Forest Service Proceedings RMRS-P-31. 2004
105
DeBolt and Spurrier
Seeds of Success and the Millennium Seed Bank Project
Table 1—(Con.)
Family
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Polemoniaceae
Polemoniaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Polygonaceae
Portulacaceae
Primulaceae
Primulaceae
Primulaceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Ranunculaceae
Rhamnaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rosaceae
Rutiaceae
Salicaceae
Salicaceae
Salicaceae
Salicaceae
Salicaceae
Santalaceae
Genus
Pluchea
Poa
Poa
Poa
Sporobolus
Stipa
Stipa
Collomia
Polemonium
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Eriogonum
Oxyria
Rumex
Lewisia
Dodecatheon
Primula
Samolus
Aconitum
Aconitum
Anemone
Aquilegia
Caltha
Clematis
Thalictrum
Rhamnus
Amelanchier
Cercocarpus
Cercocarpus
Cercocarpus
Dryas
Dryas
Fallugia
Geum
Geum
Luetkea
Potentilla
Potentilla
Prunus
Purshia
Purshia
Purshia
Rosa
Rosa
Rubus
Sanguisorba
Spiraea
Spiraea
Thamnosma
Populus
Salix
Salix
Salix
Salix
Commandra
Species
sericea
fendleriana
macrantha
secunda
airoides
comata
lettermanii
grandiflora
foliosissimus
bifurcatum
caespitosum
corymbosum
latifolium
lobbii
niveum
nudum v. auriculatum
racemosum
sphaerocephalum v. halimioides
umbellatum ssp. nevadense
umbellatum v. polyanthum
digyna
arcticus
rediviva
redolens
parryi
ebracteatus
columbianum
delphiniifolium ssp. delphiniifolium
parviflora
formosa
leptosepala
ligusticifolia
fendleri
betulifolia
alnifolia
ledifolius v. intermontanus
ledifolius
montanus
drummondii
octopetala ssp. octopetala
paradoxa
rossii
triflorum
pectinata
gracilis
uniflora
virginiana
glandulosa
mexicana
tridentata
pisocarpa
woodsii
parviflorus
canadensis
douglasii
stevenii
montana
angustifolia
exigua
lemmonii
lucida
lutea
umbellatum
State
NV, UT
NV
CA
NV
NV, UT
UT
CO
CA
UT
NV
CA
NV
CA
NV
WA
OR
UT
CA
CA
CA
AK
AK
ID
NV, UT
WY
NV
UT
AK
AK
CA
UT
UT
UT
NV
ID
NV
UT
CO
AK
AK
UT
AK
CA
AK
CA
AK
ID
NV
UT
UT
OR
CO
CA
AK
OR
AK
NV
UT
ID, UT
ID
UT
ID, UT
NV
(con.)
106
USDA Forest Service Proceedings RMRS-P-31. 2004
Seeds of Success and the Millennium Seed Bank Project
DeBolt and Spurrier
Table 1—(Con.)
Family
Genus
Anemopsis
Boykinia
Parnassia
Saxifraga
Saxifraga
Saxifraga
Tellima
Tiarella
Castilleja
Castilleja
Castilleja
Castilleja
Castilleja
Lagotis
Mimulus
Mimulus
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Penstemon
Scrophularia
Veronica
Veronica
Typha
Valeriana
Larrea
Saururaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Saxifragaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Scrophulariaceae
Typhaceae
Valerianaceae
Zygophyllaceae
Center for Plant Conservation botanists collected in two
States. Table 2 includes both target and the actual number
of taxa collected by State. In general, where interns or other
assistance was provided, targets were exceeded, as the
Table 2—2002 targeted and actual number of taxa
collected by the Bureau of Land Management
for the Millenium Seed Bank project and Seeds
of Success program.
State
AK
AZ
CAa
CO
IDa
MT
NVa
NM
ORa
UTa
WY
Total
a
Number of species
targeted
20
25
50
25
50
20
50
20
50
50
30
390
Number of species
collected
45
1
73
11
37
0
77
0
52
79
0
375
States that had Student Conservation Association intern teams.
USDA Forest Service Proceedings RMRS-P-31. 2004
Species
californica
occidentalis
kotzebuei
bronchialis
lyallii
odontoloma
grandiflora
trifoliata ssp. unifoliata
applegatei ssp. pinetorum
linariifolia
minor ssp. minor
pruinosa
unalaschcensis
minor
aurantiacus
guttatus
acuminatus
albomarginatus
cyananthus
cyaneus
deustus
fruticiformis
fruticosus
laevis
palmeri
platyphyllus
roezlii
californica
peregrina ssp. xalapensis
wormskjoldii
angustifolia
sitchensis
tridentata
State
CA, NV
OR
AK
AK
AK
UT
OR
OR
CA
CA, UT
UT
OR
AK
AK
CA
CA, UT
ID
NV
UT
ID
ID
NV
ID
UT
UT
UT
CA
CA
CA
AK
UT
AK
NV
additional workload was difficult to accomplish otherwise.
Drought conditions in the Western United States limited
seed production for some species; however, because flexibility is built into the program, alternate species could
usually be identified for collection.
Seed collecting protocols, shipping forms, proposed species lists, and a database of collected taxa are available at
http://www.nps.gov/plants/sos and http://www.rbgkew.org.
uk/seedbank. Digital photographs of each species and its
seed were also taken in 2002. These photographs are housed
at RBG, Kew, and each participating BLM office, and may
eventually be available online.
Conclusions ____________________
In 2002, in partnership with RBG, Kew, and as part of the
international Millennium Seed Bank project, the Bureau of
Land Management implemented a native seed collecting
program deemed “Seeds of Success.” Seed from 414 populations representing 375 taxa were collected for long-term
storage at RBG, Kew, and the USDA National Seed Storage
Laboratory in Fort Collins, CO, and for use in native plant
materials development. As a multiple-use agency, the BLM
has a need for native plant materials for many different
programs in order to meet its goal of restoring at-risk
107
DeBolt and Spurrier
resources and maintaining or improving functioning ecosystems. The Seeds of Success program facilitates development
of native plant materials from local genetic sources, and
provides valuable information on germination and viability
of native plant species that are currently in short supply.
Additional information on Seeds of Success and the Millennium Seed Bank project is available at http://www.nps.gov/
plants/sos and http://www.rbgkew.org.uk/seedbank.
108
Seeds of Success and the Millennium Seed Bank Project
References _____________________
Brown, A. H. D.; Marshall, D. R. 1995. A basic sampling strategy:
theory and practice. In Guarino, L; Ramanatha Rao, V.; Reid, R.
Collecting plant genetic diversity. Technical Guidelines. CAB
International.
Kartesz, J. T.; Meacham, C. A. 1999. Synthesis of the North America
Flora, Version 1.0. North Carolina Botanical Garden, Chapel
Hill, NC.
USDA Forest Service Proceedings RMRS-P-31. 2004
Recruitment Failure of Chenopod
Shrub Populations in the Great
Basin
Susan C. Garvin
Susan E. Meyer
David L. Nelson
Abstract: The salt desert shrub ecosystem has suffered increasing
invasion by exotic annual weeds, especially cheatgrass, for the last
25 years or longer. After the extensive shrub dieoffs of 1982 to 1984
in the Great Basin and Colorado Plateau, many shadscale (Atriplex
confertifolia) and winterfat (Ceratoides lanata) populations failed to
regenerate. Several facultatively pathogenic fungi were isolated
from the rhizosphere of shadscale. Levels of soil Fusarium and
pythiaceous fungi in intact and dieoff shrubland areas were measured with positive correlation between fungal density and dieoff
severity of shadscale in one study and winterfat in a separate study.
Shadscale seedings in three shadscale sites invaded by annual
weeds in the Utah western desert resulted in significant seedling
emergence followed by 100 percent establishment failure. Seedlings
with wilt symptoms were observed in each of these studies. In each
case soil moisture was adequate for seedling establishment. A
greenhouse test was conducted to study seedling establishment of
these two species in 10 valley bottom soils, four from recruiting
shrub populations and six from defunct shrub sites invaded by
cheatgrass or halogeton. Autoclaved and fungicide-treated soil
treatments significantly increased winterfat seedling survival (P >
0.001) in soils from weed-invaded sites. No effect of soil treatment
was seen in shadscale seedling numbers in this study. Winterfat
seedling weight was not affected by soil treatment and did not differ
between soil sites. Shadscale seedling weight was significantly
increased by soil treatment in intact soils but not in weedy soils. In
a separate emergence study shadscale fruits were planted in soil
from two intact and two weed-invaded shadscale sites. After moist
chilling at 2 ∞C, more than twice as many live seedlings were
observed in the two soils from intact sites as in the two soils from
weed-invaded sites. Work in progress will attempt to isolate a causal
organism from dying shadscale seedlings. Shadscale sensitivity to
fungicides will also be investigated.
over 17 million ha of lowlands, with the dominant perennial
plants being shrubs from the chenopod family (West 1994).
During the last 25 to 50 years increasing invasion of alien
grasses and other annual plants has led to recurring fires
and expanding invasion of weeds; in many cases the native
shrub communities have not returned.
An extensive shrub dieoff occurred in the Great Basin and
Colorado Plateau in the early 1980s, coinciding with a series
of “El Niño” years that resulted in record high precipitation.
This dieoff affected chenopod shrubs, particularly shadscale
(Atriplex confertifolia) in most Great Basin valleys. The
Bureau of Land Management (BLM) estimated that more
than 280,000 ha of shadscale were lost in Utah alone (Nelson
and others 1989). This dieoff was followed by a massive
invasion of annual weeds and a subsequent series of wildfires. Shadscale recruitment in most burned dieoff areas was,
and remains, almost nonexistent. It is common for shadscale
populations to experience cyclic dieoff and recruitment events,
made possible by a long-lived soil seed bank (Sharp and
others 1990), but the recruitment phase of that cycle has
failed to occur in many populations, and shadscale seed
banks may have become depleted. Widespread decline and
recruitment failure of winterfat (Ceratoides lanata), a shrub
second only to shadscale in abundance in salt desert shrubland, has also occurred in many Great Basin valleys (Kitchen
2001; Nelson and others 1989). Thus substantial decreases
of two major shrubs have resulted in large acreages of valley
bottoms susceptible to invasion by cheatgrass (Bromus
tectorum), halogeton (Halogeton glomeratus), and other
weeds.
Introduction ____________________
This paper will review published studies explaining the
causes of this phenomenon and describe more recent preliminary studies carried out in our laboratory. Studies
carried out during and shortly after the 1980s El Niño
investigated, among other factors, the role of pathogenic
fungi in the dieoff. Shadscale roots were exhumed, and
fungal isolations were made from surface-sterilized tap and
secondary roots, as well as crowns and stems (Nelson and
others 1990). Hundreds of fungi were isolated, pure-cultured, and identified to genus and often species level. Fungi
with pathogenic potential were frequently isolated from
surface sterilized root material. Fusarium was the most
frequently isolated genus, from both healthy and necrotic
The salt desert shrub ecosystem in Western North America
is second in area only to the sagebrush ecosystem. It covers
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Susan C. Garvin is Biological Technician, Susan E. Meyer is Ecologist, and
David L. Nelson is Retired Plant Pathologist, U.S. Department of Agriculture
Forest Service, Shrub Sciences Laboratory, Provo, UT, U.S.A. FAX: (801)
375-6968; e-mail: sgarvin@fs.fed.us
USDA Forest Service Proceedings RMRS-P-31. 2004
Previous Research ______________
109
Garvin, Meyer, and Nelson
Recruitment Failure of Chenopod Shrub Populations in the Great Basin
root material and at every plant position. Facultatively
parasitic plant pathogens are common in this genus; this
class of fungi can survive and multiply saprophytically on
organic substrates, but can also induce plant disease, especially when plant host defense mechanisms are lowered
because of environmental stress (Nelson and others 1990).
Some Fusarium species commonly induce both vascular wilt
and root rot in susceptible plants. Other genera commonly
isolated were Alternaria, known to cause leaf spot and twig
blight, Cephalosporium, also capable of causing vascular
wilt, and Sclerotium, Rhizoctonia, and Pythium, all of which
can cause root rot or “damping-off” of seedlings.
As a part of these dieoff studies, soil samples were
collected from the rhizosphere of shadscale and other
dying shrubs, and populations of fungi closely related to
Pythium (referred to hereafter as pythiaceous fungi) were
counted on selective agar media. A very strong correlation
between small secondary rootlet mortality and dieoff severity (r = –0.99, P = <0.01) and a significant correlation
between number of pythiaceous fungi and dieoff severity
on a site (r = 0.64, P = <0.01) was observed (fig. 1). The
number of pythiaceous propagules was quite variable and
ranged from 163 to 2,931 per gram of soil in sites with a
low dieoff severity to 1,416 to 6,056 per gram of soil in sites
with complete shadscale dieoff.
Harper and others (1996) reported on a winterfat dieoff
occurring on the Desert Experimental Range in southwestern
Utah. Winterfat was being replaced by halogeton in a valley
bottom, the site of a known overland waterflow. In an adjacent
site winterfat was recruiting successfully as openings occurred. Soil samples were collected from both the winterfat
dieoff area and adjacent healthy winterfat area; half the soil
was fumigated with methyl bromide, and winterfat seeds
6
4
4
2
2
Live rootlets x 10
1
6
were sown in pots in the greenhouse. Results from that study
(Harper and others 1996) showed that fumigation increased
winterfat recruitment in halogeton-affected soil but not in soil
from the healthy winterfat site.
Further work on these soils carried out in our laboratory
was designed to determine if density of pathogenic fungi
correlated with lack of winterfat recruitment. Surface soil
samples were collected in three areas: dieoff areas replaced
by halogeton, dieoff areas not yet invaded by halogeton, and
areas where winterfat was successfully recruiting. Soil
samples were diluted and plated on two agar media, selective for either Fusarium species or pythiaceous species, and
number of colonies was determined, with each colony representing a reproductive propagule. In winterfat dieoff soils
Fusarium numbers were 3.4 times higher than in soils from
intact winterfat areas (P > 0.001) (table 1). Pythiaceous
fungal numbers were more than 10 times higher in winterfat
dieoff soil than in intact winterfat soil (P > 0.001). This trend
continued and intensified in areas where winterfat had been
replaced by halogeton (David L. Nelson, unpublished data).
Ramsey and others (1995) presented results of a study
conducted in 1992 to 1994 on an Atriplex seedling failure in
northeastern New Mexico. A selection of an Atriplex hybrid
of A. canescens X A. obovata repeatedly failed to become
established after seeding in a nonmined site. Soil from the
site and a commercial soil mix were planted with the same
seed source under greenhouse conditions, and establishment was compared. Seedlings emerged in both soils, but
most of the seedlings in the native soil died of “damping-off.”
Most seedlings in the commercial mix survived.
Diluted soil samples from the planting site and a nearby
stockpile soil were plated on selective agar media designed
to inhibit all but pythiaceous fungi. The planting site soil
averaged 1,230 pythiaceous fungi per gram of soil, compared
to 20 pythiaceous fungi per gram from the stockpile soil.
Seeds treated with the fungicide Captan® (Drexel Chemical
Co.) and untreated seeds were planted in a pot experiment.
Seedling mortality in the soil from the failed planting site
was 87 percent using untreated seed and 27.5 percent with
the Captan® treated seed. A subsequent seeding at the same
site using Captan®-treated seeds increased seedling survival 100-fold.
New Research __________________
0
0
0
1
2
3
4
5
6
7
Shadscale Seedling Establishment
Studies
In the course of our studies on shadscale germination
biology (Garvin and others 1995; Garvin and Meyer 2003;
Dieoff severity index
Figure 1—Correlation of dieoff severity at
shadscale sites with number of live rootlets present
on 30 plants excavated from each severity class
and with number of pythiaceous fungi isolated
from rhizosphere soil of 10 plants from each of 20
sites. Each site was ranked by averaging plant
dieoff class (1 to 6) at 10-m intervals along three
100-m transects, so a site with dieoff severity of 6
consisted of all dead plants (Nelson and others
1990).
110
Table 1—Enumeration of Fusarium and pythiaceous fungi in soil from
a winterfat dieoff site in western Utah. Numbers are fungal
propagules per gram of soil.
Soil source
Healthy winterfat area
Winterfat dieoff area
Halogeton invaded dieoff area
Fusarium
517
1,771
4,775
Pythiaceous fungi
105
1,152
1,765
USDA Forest Service Proceedings RMRS-P-31. 2004
Recruitment Failure of Chenopod Shrub Populations in the Great Basin
Garvin, Meyer, and Nelson
Meyer and others 1998), we carried out several field tests to
evaluate seedling emergence and establishment. These studies were carried out in two western Utah valleys in 1991,
1997, and 2000. The 1991 study was conducted in northeastern Rush Valley with eight populations of shadscale fruits in
emergence strips 8 ft long by 1 ft wide; each population was
replicated eight times. A total of approximately 19,200 filled
fruits were planted 1 cm deep. The study in 1997 consisted
of four populations, planted in square-meter plots, replicated eight times, with approximately 32,000 filled fruits
planted. A third study was carried out in 2000 using seed
already in the seed bank, so total fruit density is not known.
Emergence and establishment results are shown in table 2.
In each test, a considerable fraction of seeds germinated and
emerged; 3 months later only a small fraction of emerged
seeds survived, and by the following spring none of the
seedlings remained alive. Each time we counted seedlings
we observed typical damping-off symptoms, even though soil
moisture was adequate for survival.
Greenhouse Seedling Establishment of
Winterfat and Shadscale
In 1999 we began a preliminary greenhouse pot study to
determine the effect of different soils on seedling establishment. We collected soil (upper 5 to 10 cm) from each of 10
sites in four valley bottoms in western Utah, attempting to
match weed-invaded sites with equivalent intact sites. Soil
was either autoclaved to eliminate all microorganisms,
treated with Banrot® (The Scotts Company) as a drench at
the rate of 12 ounces per 100 gallons) or untreated. Rectangular pots (10 cm by 15 cm by 7.5 cm) were filled with soil,
watered to saturation, and planted either with 25 winterfat
seed or 25 shadscale seedlings germinated in Petri dishes in
moist chill. Each soil treatment x species combination was
replicated eight times. Seedling emergence was evaluated
every 3 days initially and once a week after the first 2 weeks.
After 3 months, plants were harvested by clipping at the soil
line, dried, and weighed. We compared weedy and intact
soils and different soil treatments using SAS PROC GLM
Analysis of Variance.
The effect of fungicide treatment on plant survival and
size is shown in figure 2. Autoclaved soil treatment results
are similar. Winterfat seedling survival was significantly
higher in fungicide treated and autoclaved soil treatments
from weedy sites, but little effect was seen on survival in soil
from intact sites. Most death of winterfat seedlings occurred
10 to 21 days after emergence with little additional seedling
death after that point. We found no significant effect of soil
Table 2—Number of seedlings emerged in early spring, established in
late spring, and established 1 year later after a late fall
planting.
Rush Valley
Skull Valley North
Skull Valley South
Early spring
seedlings
Late spring
seedlings
807
1,029
896
118
24
114
USDA Forest Service Proceedings RMRS-P-31. 2004
Year-old
plants
treatment on shadscale seedling numbers. Percentage survival of transplants was similar in all three soil treatments.
Little difference in plant weight of winterfat was measured overall, although in most soils either autoclaving or
drenching (but often not both) produced significantly greater
plant growth than in untreated soil. The ratio of shadscale
plant weight in fungicide treated soil compared to untreated
soil from nonweedy sites was significantly increased by a
factor of 2 (P > 0.001). No significant effect of fungicide or
autoclave treatment was seen on shadscale plant weight
when plants were grown in weedy soil.
Shadscale Seedling Survival During Moist
Chilling
A second preliminary study was carried out in a cold room
and growth chamber. This study used four soil collections, two
from a weed-invaded site and two from a nearby intact site in
two western Utah valleys. Four different methods of fungus
control, including biological control organisms, were tested.
Subdue® (Ciba-Geigy) fungicide was applied at the recommended high rate as a drench; soil was steamed at 60 ∞C for
30 minutes; Kodiak® (commercial Bacillus subtilis GBO3
inoculum obtained from Gustafson) was applied to fruit
before planting; and commercial Gliocladium inoculum
(Soilguard®) from Gardens Alive! was pre-applied to fruit in
combination with a half dosage of Subdue® fungicide applied
as a soil drench. Afterripened shadscale fruits were planted in
the rectangular pots as described in the previous study,
watered to saturation, placed in closed plastic bags to maintain soil saturation, and incubated at 2 ∞C for 12 weeks.
Treatments were replicated 15 times. Pots were removed
from the cold, germinated seedlings were counted in each pot,
and pots were moved to the growth chamber, set to approximate early spring temperature and light conditions. Unfortunately, the experiment was terminated prematurely due to
overheating of the growth chamber and death of most seedlings, so only main effect means were analyzed statistically.
Results of the seedling count are summarized in table 3.
This data is a measure of seedling emergence and survival
during moist chilling. The steamed soil treatment is not
shown since almost no germination occurred in any treatment. There was a strong indication in our data that shadscale
seedlings may have been damaged before emergence by the
high rate of Subdue® fungicide, since seedling emergence in
the fungicide treated soil was only 31.7 percent of the control,
and only 56 percent of emerged seedlings survived, compared
with 71 to 76 percent survival in the other soil treatments.
Emergence of plants grown in the soil treated with only onehalf the high fungicide dosage (with SoilGuard®) was significantly less than emergence of plants grown in the untreated
soil (P > 0.05); plant survival after emergence was not significantly reduced by the half dosage of fungicide. More than
twice as many seedlings emerged and survived in soils collected from intact shadscale sites as in soils from weedinvaded sites (P > 0.01). No other differences between fungus
control treatments could be determined.
0
2
0
111
Garvin, Meyer, and Nelson
Recruitment Failure of Chenopod Shrub Populations in the Great Basin
2.5
3.0
Intact site
Weedy site
Ratio drenched/untreated
seedling number
Ratio drenched/untreated
seedling mass
3.0
2.0
1.5
1.0
0.5
0.0
2.5
2.0
1.5
1.0
0.5
0.0
CELA
ATCO
CELA
ATCO
Figure 2—Means of the ratio of fungicide drench weight or seedling number to untreated values for soils
from four intact sites versus soils from six weedy sites in the greenhouse pot experiment. Larger ratios
indicate more effect of fungicide, with values of 1.0 indicating no effect.
Discussion _____________________
Because shadscale seed was not germinated in the presence of soil fungi in the greenhouse study, it is not surprising
that seedling numbers were little affected by soil treatment.
Damping-off organisms are often attracted to germinating
seeds (Schroth and Cook 1964) so seedlings that might have
become infected in the test soils during this period survived.
Pathogens in the test soils seldom killed shadscale seedlings
after emergence. Increased survival of winterfat seedlings
in treated weedy soils can be interpreted to mean that a
group of pathogenic fungi was suppressed by fungicide and
autoclaving. The interpretation of the shadscale weight data
becomes more problematic. If pathogens were responsible
for inhibited plant growth we would expect that soil treatments that decrease numbers of facultative soil fungi would
increase plant weight. We found this to occur in the soil from
intact shadscale sites but surprisingly not in the soil from
weedy sites. One interpretation might be that fungicide
treatment suppressed endemic low levels of pathogens (root
nibblers) and promoted more growth in intact soils but was
ineffective against a different pathogen or pathogens in
weedy soil. These arguments lead to the possibility of several
different pathogens affecting chenopod shrubs, in both weedinvaded and intact soils, and at different stages of plant
growth. Other evidence for this scenario comes from Nelson’s
(1990) studies, which show the presence of several facultative fungal pathogens in root and stem tissue of shadscale
plants. It is difficult to test for pathogenicity of these organisms because plants must be stressed in some way to lead to
the infection process.
Some likely confounding effects of autoclaving soil include
the production of toxic organic compounds during heating
and the increase in available nutrients due to the death of
Table 3—Mean fraction of shadscale seedling survival and number of emerged seedlings (live and dead) after 2 months in moist chilling at 2 ∞C
in pots with soils from two intact and two weedy shadscale sitesa.
Soil source
Fraction of surviving
shadscale seedlings
Dugway intact
Rush intact
Dugway weedy
Rush weedy
Treatment means
Number of emerged
seedlings
Dugway intact
Rush intact
Dugway weedy
Rush weedy
Treatment means
a
b
112
Untreated
Fungicide
0.671
.658
.850
.658
.709ab
0.588
.687
.609
.357
.560b
70
117
40
38
66.25a
17
30
23
14
21.00c
Kodiak®
0.690
.861
.682
.628
.715ab
71
79
22
43
53.75ab
SoilGuard®+1/2 Fb
0.702
.773
.792
.778
.761a
47
44
24
18
33.25bc
Site means
0.663
.745
.733
.605
51.25
67.50
27.25
28.25
Weediness means
.704a
.669a
59.38a
27.75b
Main effect means followed by the same letter are not significantly different at the P < 0.05 level (LSD).
1/2 F = One-half the full dosage of fungicide. See text for details on soil treatments and sources of biological control treatments.
USDA Forest Service Proceedings RMRS-P-31. 2004
Recruitment Failure of Chenopod Shrub Populations in the Great Basin
the microbial population; these factors were not controlled in
our test and probably explain some of the variability between soils and inconsistent results within soils. Another
disadvantage of steaming or autoclaving soils high in clay
content is apparent in our study on seedling emergence
during moist chilling: the heated soils lost all structure and
apparently became either toxic or impenetrable to seedlings.
Soil fumigation might lead to clearer results, although
fumigation, as well as autoclaving, can cause a flush of
available nitrogen after soil biota are killed.
Another confounding factor is that fungicide treatment of
soil may be toxic to some plant species (Leach and MacDonald
1978). If the clay fraction of the soil is lower than in agricultural soil, less fungicide is bound by the clay fraction.
Conversely, if the clay fraction is higher than in most
agricultural soils, more fungicide may be bound by clay
particles and not available for killing fungal pathogens. Soil
organic matter can also immobilize fungicide, so increased
organic matter in weed-infested soils may bind more fungicide than in soils from intact shrub sites. More research
needs to be carried out to discover effective rates of application of fungicide on chenopod shrubs and in different desert
soils. Because we do not know which fungi are acting as
pathogens, we may not have used an appropriate fungicide.
We are currently engaged in studies on shadscale seeds
designed to isolate possible pathogens from infected shadscale
seedlings. Candidate fungi will then be tested by inoculating
shadscale fruits and re-isolating the fungus from infected
seedlings. We will also determine appropriate rates of fungicide application for the valley bottom soils from our study
sites and test these rates for toxicity to shadscale germination and seedling stages.
The massive chenopod shrub dieoff associated with record
precipitation and subsequent invasion of annual weeds has
also resulted in changes in the soil. Harper and others (1996)
measured increased salinity in the surface soil of halogetoninfested winterfat sites. Increased salinity probably cannot
account for the death of mature shrubs, but plants in the
germination and young seedling stage are far more susceptible to increased salinity, so this could be a factor in
recruitment failure.
Ewing and Dobrowolski (1992) characterized soil factors
such as moisture and salinity in the plant communities in
Puddle Valley, UT, where shadscale dieoff was extensive in
the early 1980s. They found that dead shadscale was negatively correlated with slope and elevation; the greatest
density of dead shrubs was found in flat valley bottoms.
Shadscale death was positively correlated with increased
clay content and increased soil moisture, both conditions
leading to enhancement of favorable sites for fungus proliferation. Sites with higher salinity had greater rates of
survival. The leading edge of shadscale dieoff was followed
from 1987 to 1989; the authors concluded that the dieoff was
continuing even after precipitation levels returned to normal.
Topsoil organic matter and surface litter also increase in
annual weed-infested sites, which is likely to increase the
populations of facultatively parasitic fungi, such as Fusarium
and Pythium. Other soil biotic changes may also occur, as
shown by Evans and Belnap (1999), such as decreased nitrogenase and ammonia mineralization activity in sites with
disturbed cryptobiotic crust. Belnap and Phillips (2001) found
that plant pathogenic fungi increased in cheatgrass-invaded
USDA Forest Service Proceedings RMRS-P-31. 2004
Garvin, Meyer, and Nelson
perennial grass sites in Canyonlands National Park. Bolton
and others (1993) measured increased surface soil microbial
biomass carbon and nitrogen, soil respiration, dehydrogenase, and phosphatase activity in cheatgrass invaded soil
compared to soil under sagebrush, perennial grass, and
cryptobiotic crust. The changes to soil biota and chemistry
caused by annual weed invasion have only begun to be
studied; effective measures to restore native communities
need to consider changes within as well as above the soil.
References _____________________
Belnap, J.; Phillips, Susan L. 2001. Soil biota in an ungrazed
grassland: response to annual grass (Bromus tectorum) invasion.
Ecological Applications. 11: 1261–1275.
Bolton, H., Jr.; Smith, J. L.; Link, S. O. 1993. Soil microbial biomass
and activity of a disturbed and undisturbed shrub-steppe ecosystem. Soil Biology and Biochemistry. 25: 545–552.
Evans, R. D.; Belnap, J. 1999. Long-term consequences of disturbance on nitrogen dynamics in an arid ecosystem. Ecology. 80:
150–160.
Ewing, K.; Dobrowolski, J. P. 1992. Dynamics of shrub die-off in a
salt desert plant community. Journal of Range Management. 45:
194–199.
Garvin, S. C.; Meyer, S. E.; Carlson, S. G. 1996. Seed germination
studies in Atriplex confertifolia (Torr. & Frem.) Wats. In: McArthur,
E. D.; Barrow, J. R.; Sosebee, R. E.; Tausch, R. J., comps.
Proceedings—shrubland ecosystem dynamics in a changing environment; 1995 May 23–25; Las Cruces, NM. Gen. Tech. Rep. INTGTR-338. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station: 165–169.
Garvin, S. C.; Meyer, S. E.; Clement, S. 2003. Multiple mechanisms
for seed dormancy regulation in shadscale (Atriplex confertifolia:
Chenopodiaceae). Canadian Journal of Botany. 81: 601–610.
Harper, K. T.; Van Buren, R.; Kitchen, S. G. 1996. Invasion of alien
annuals and ecological consequences in salt desert shrublands of
western Utah. In: Barrow, J. R.; McArthur, E. D.; Sosebee, R. E.;
Tausch, R. J., comps. Proceedings—shrubland ecosystem dynamics in a changing environment; 1995 May 23–25; Las Cruces, NM.
Gen. Tech. Rep. INT-GTR-338. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain Research Station:
58–65.
Kitchen, S. 2001. Winterfat decline and halogeton spread in the
Great Basin. In: McArthur, E. D.; Fairbanks, D. J., comps.
Proceedings—shrubland ecosystem genetics and biodiversity;
2000 June 13–15; Provo, UT. Gen. Tech. Rep. RMRS-P-21. Ogden,
UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 200–203.
Leach, L. D.; MacDonald, J. D. 1978. Greenhouse evaluation of seed
treatment fungicides for control of sugar beet seedling diseases.
In: Zehr, Eldon I., ed. Methods for evaluating plant fungicides,
nematicides, and bactericides. St. Paul, MN: American Phytopathological Society: 23–25.
Meyer, S. E.; Carlson, Stephanie L.; Garvin, Susan C. 1998. Seed
germination regulation and field seed bank carryover in shadscale
(Atriplex confertifolia: Chenopodiaceae). Journal of Arid Environments. 38: 255–267.
Nelson, David L.; Harper, Kimball T.; Boyer, Kenneth C.; Weber,
Darrell J.; Haws, B. Austin; Marble, James R. 1989. Wildland
shrub dieoffs in Utah: an approach to understanding the cause.
In: Proceedings—symposium on shrub ecophysiology and biotechnology; 1987 June 30–July 2; Logan UT. Gen. Tech. Rep. INT276. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Intermountain Research Station: 119–135.
Nelson, D. L.; Weber, D. J.; Garvin, S. C. 1990. The possible role of
plant disease in the recent wildland shrub dieoff in Utah. In:
McArthur, E. Durant; Romney, Evan M.; Smith, Stanley D.;
Tueller, Paul T., comps. Proceedings—symposium on cheatgrass
invasion, shrub die-off, and other aspects of shrub biology and
management; 1989 April 5–7; Las Vegas, NV. Gen. Tech. Rep.
INT-276. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station: 84–90.
113
Garvin, Meyer, and Nelson
Ramsey, T.; Stutz, H. C.; Nelson, D. 1995. Seedling failure in
Atriplex (Chenopodiaceae) due to “damping-off.” In: Schuman,
Gerald E.; Vance, George F., eds. Proceedings of national meeting
of the American Society for Surface Mining and Reclamation;
1995 June 5–8, Gillette, WY. Princeton, WY: American Society for
Surface Mining and Reclamation: 757–761.
Sharp, L. A.; Sanders, K.; Rimbey, N. 1990. Forty years of change in
a shadscale stand in Idaho. Rangelands. 12: 313–328.
114
Recruitment Failure of Chenopod Shrub Populations in the Great Basin
Schroth, M. N.; Cook, R. J. 1964. Seed exudation and its influence
on pre-emergence damping-off of bean. Phytopathology. 54:
670–673.
West, Neil E. 1994. Effects of fire on salt-desert rangelands. In:
Monsen Stephen B., Kitchen, Stanley G., comps. Proceedings—
symposium on ecology, management, and restoration of Intermountain annual rangelands; 1992 May 18–22; Boise ID. Gen.
Tech. Rep. INT-313. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Research Station: 71–74.
USDA Forest Service Proceedings RMRS-P-31. 2004
Annual Brome Seed Germination
in the Northern Great Plains: An
Update
M. R. Haferkamp
M. D. MacNeil
Abstract: Japanese brome (Bromus japonicus Thunb) and downy
brome (B. tectorum L.), alien weedy cool-season annual grasses,
have invaded thousands of hectares of grass and shrub communities
in the Northern Great Plains, Great Basin, and Columbia Basin.
Abundance of brome is dependent upon availability of seed, amount
and distribution of rainfall, temperature, and availability of soil
nitrogen. More than 10,000 annual brome seeds can be present in a
square meter in the mixed-grass prairie of the Northern Great
Plains. A large portion of ripe seeds will generally germinate over a
wide range of temperatures and osmotic potentials that often occur
in late summer and autumn, but soils usually must be moist for 3 to
5 days for seeds to germinate. However, a percentage of seeds that do
not germinate by late September can become dormant when water is
taken up at or below 0 ∞C. This dormant state can last through the
next winter, spring, and summer. These characteristics aid annual
brome’s persistence on rangelands.
Introduction ____________________
Japanese brome (Bromus japonicus Thunb) and downy
brome (B. tectorum L.), alien weedy cool-season annual
grasses, have invaded thousands of hectares of grass and
shrub communities in the Northern Great Plains (Haferkamp
and others 1993; Hewlett and others 1981; Whisenant 1990),
Great Basin (Mack 1981; Young and Evans 1972), and
Columbia Basin (Daubenmire 1970; Hulbert 1955). Although
annual bromes can produce large quantities of nutritious
spring forage, they can have a negative impact on associated perennial forage species and performance of grazers
(Haferkamp and others 1994b, 1997, 1998, 2001a,b;
Rummell 1946). Forage production is erratic from year to
year (Gartner and others 1986; Haferkamp and others
1993, 2001a), and due to early maturation of the annual
bromes (Vallentine and Stevens 1994) compared to perennial grasses, their presence can alter seasonal patterns of
forage production and cause relatively rapid declines in
forage quality on infested rangelands (Haferkamp and
others 1994b, 2001b).
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
M. R. Haferkamp is a Rangeland Scientist and M. D. MacNeil is a Research
Geneticist, USDA-ARS, Fort Keogh Livestock and Range Research Laboratory, 243 Fort Keogh Road, Miles City, MT 59301-4016, U.S.A., FAX: (406)
232-8209, email: marshall@larrl.ars.usda.gov
USDA Forest Service Proceedings RMRS-P-31. 2004
Abundance of brome is dependent upon the seedbank
(Karl and others 1999), temperature, and amount and distribution of rainfall (Haferkamp and others 1993; Mack and
Pyke 1983; Whisenant 1990). Karl and others (1999) reported that more than 10,000 germinable annual brome
seeds can be present in a square meter in the mixed-grass
prairie of the Northern Great Plains. Thus, the potential for
dense populations of annual brome seedlings is great. We
have developed a better understanding of the dynamics of
seed germination for Japanese brome through a series of
studies conducted at the Fort Keogh Livestock and Range
Research Laboratory, Miles City, MT (Haferkamp and others 1994a, 1995a,b).
Temperature ____________________
The influence of temperature on germination was investigated on Japanese brome seed collected on native rangeland
in eastern Montana (Haferkamp and others 1995b). The
objective was to determine temperature profiles for germination of Japanese brome seeds, representative of those
disseminated from seed heads during the period spanning
late summer to early autumn.
Methods
Seeds were collected on eight dates during July and
August 1990 and 1991. Seeds were stored in paper sacks in
the laboratory at about 21 ∞C until germination trials were
run in Reno, NV, 3 to 9 months after collection.
Four replications of 25 seeds from each collection date
were incubated for 4 weeks in closed petri dishes on 1-mmthick filter paper moistened with distilled water. Dishes
were arranged in a randomized complete-block design, in 10
dark incubators. Constant incubator temperatures included
0, 2, 5, and all 5 ∞C increments through 40 ∞C. Alternating
temperature regimes were attained by moving dishes between incubators using 16 hours at each constant temperature plus 8 hours at all possible higher temperatures. For
example, 35 ∞C was alternated with 40 ∞C, while 0 ∞C was
alternated with 2, 5, 10, 15, 20, 25, 30, 35, and 40 ∞C. Germination counts were taken weekly, and seeds were considered
germinated when the radicle was at least 5 mm long.
Data from each base temperature and its alternating
temperature regime were used to develop regression equations of quadratic response surfaces with estimated germination means and confidence intervals at the 5-percent level
of probability. A number of germination parameters were
115
Haferkamp and MacNeil
Annual Brome Seed Germination in the Northern Great Plains: An Update
synthesized from these quadratic response surfaces to assist
in interpretation of germination temperature profiles.
Results and Discussion
Mean germination of Japanese brome seed from the 55
temperature profiles averaged across the eight collection
dates represented in figure 1 was 71 + 4.0 percent. Some
germination occurred in 96 + 1.4 percent of the temperature regimes, and the mean germination for regimes where
some germination occurred was 74 + 4.0 percent. Optimum
temperatures for germination of Japanese brome seed
were constant 20 ∞C and several alternating (16 hours of
minimum with 18 hours of maximum) temperatures. The
optimum alternating regimes were 2 through 10 ∞C minimums with 15 ∞C maximum, 0 through 15 ∞C minimums
with 20 ∞C maximum, 5 through 20 ∞C minimums with 25 ∞C
maximum, and 15 ∞C minimum with 30 ∞C maximum. Over
75 percent of the collections produced optimum germination (defined as germination not lower than the maximum
observed minus one-half of its confidence interval at 0.05
level of probability) in these regimes. Maximum germination of Japanese brome seed occurred at moderate and cold
seedbed conditions with germination being somewhat depressed in very cold and warmer than moderate temperatures. Young and Evans (1985) report that at least some
germination of downy brome seed is possible with temperatures as low as 0 ∞C and as high as 40 ∞C.
A wide range of temperature conditions occur in autumn
on seedbeds in the Northern Great Plains region. Findings
of this study suggest that while Japanese brome seeds
collected in Montana are sensitive to temperature, germination in excess of 50 percent occurs over a wide range of
temperature regimes. Thus, there is an excellent potential
for continued invasion of this species on these Northern
Warm period 8 hr ∞C
0
0
2
5
10
15
20
25
30
35
40
Very cold
Cold
Cold fluctuating
2
Fluctuating
Cold period 16 hr ∞C
5
10
15
Moderate
20
25
30
35
40
Figure 1—Temperature regimes reflecting seedbed environments occurring across a wide array
of geographic locations. Regimes are based on
monitoring studies.
116
Great Plains rangelands. The high level of germination
exhibited by afterripened Japanese brome seeds suggest
that a large portion of the disseminated seeds will germinate
completely with available water during late summer and
early autumn.
Secondary Dormancy ____________
We previously collected data that suggested up to 72
percent of Japanese brome seeds disseminated after midSeptember in eastern Montana may enter a secondary
dormancy state when they imbibe water at or near 0 ∞C
(Haferkamp and others 1994a). These findings agree with
Baskin and Baskin (1981) for Japanese brome in northcentral Kentucky and with Hull and Hansen (1974) for
downy brome in Utah and Idaho. We further investigated
the potential for onset of secondary dormancy in Japanese
brome (Haferkamp and others 1995a) by designing a study
to determine which Japanese brome seed collections obtained from mid-June to late January were receptive to
induction of secondary dormancy when imbibed at 0 ∞C.
Methods
Japanese brome seeds, collected on 17 dates from June 18,
1992, to January 28, 1993, were hand threshed and stored in
near ambient outside conditions in a warehouse. In early
February 1993, eight replications of 100 seeds of each collection were incubated in 9-cm petri dishes. Each dish contained two pieces of a thick medium speed filter paper with
high water retention supported on a polyurethane foam disc.
Distilled water was supplied to the paper (Haferkamp and
others 1994a). Seeds were incubated initially for 10 days in
alternating 12-hour periods of 0 and 10 ∞C and then transferred and incubated for an additional 18 days in alternating
12-hour periods of 8 and 23 ∞C. Light was supplied with cool–2
–1
white fluorescent bulbs (PAR = 30 mmoles m sec ) during
each 12-hour 10 or 23 ∞C period. One set of seeds, the wet
regime, was moistened on day 1, and another set, the dry
regime, was not moistened until day 11, when seeds were
transferred to the 8 and 23 ∞C temperature regime. Germinated seeds were counted on days 11, 14, 21, and 28, and
removed from the dishes when the radicle and coleoptile
were at least 5-mm long. Dishes were arranged in a randomized block design. Analysis of variance was used to detect
differences in total germination data resulting from varying
seed collection dates and moisture regimes. Separate analyses were conducted for each count date. Means were separated by Least Significant Difference (P > 0.05).
Results and Discussion
Warmer
No seeds had germinated by day 11 (table 1). By day 14
less than 5 percent of the seeds germinated in the dry
regime, but up to 69 percent germinated in the wet regime.
Seed collected from June to January germinated readily in
the dry regime by day 21 (92 percent) and day 28 (97
percent), and seed collected from June through mid-September germinated 94 percent by day 21 and 96 percent by day
28 in the wet regime. Germination, however, declined in the
wet regime beginning with the late September collection to
USDA Forest Service Proceedings RMRS-P-31. 2004
Annual Brome Seed Germination in the Northern Great Plains: An Update
Haferkamp and MacNeil
Table 1—Total germination of Japanese brome seeds (includes all but moldy seeds) collected from
June 1992 to January 1993 near Miles City, MT, U.S.A., and incubated for 10 days at 0 and
10 ∞C and 18 days at 8 and 23 ∞C. Dry seeds were moistened on day 11 and wet seeds were
moistened on day 1.
Collection
date
14
Dry
Wet
Count daysa
21
Dry
Wet
28
Dry
Wet
- - - - - - - - - - - - - - - - - - - - - - - percent- - - - - - - - - - - - - - - - - - - - - - - - - Brown seed
6/18/92
0.0Aa
39.7Bc
98.4Aa
97.7Ab
99.2Aa
98.5Aa
Green seed
6/19/92
7/02/92
7/15/92
7/30/92
8/13/92
8/27/92
9/10/92
9/24/92
10/08/92
10/22/92
11/05/92
11/19/92
12/02/92
12/17/92
12/31/92
1/28/93
.0Aa
.0Aa
1.0Aa
.0Aa
.8Aa
4.0Aa
2.5Aa
.5Aa
.8Aa
.0Aa
.0Aa
.0Aa
.0Aa
.0Aa
.0Aa
.0Aa
19.3Befg
68.3Ba
69.4Ba
57.6Bb
63.2Bab
56.4Bb
27.5Bcde
19.5Bdefg
12.8Befgh
6.0Afgh
12.3Aefgh
.6Ah
4.6Afgh
6.5Afgh
3.5Agh
4.1Afgh
90.1Aab
97.5Aab
99.4Aab
96.7Aab
97.7Aab
98.8Aa
98.8Aa
97.9Aa
98.2Aa
95.9Aab
90.3Aab
69.4Ac
75.3Ac
88.0Ab
89.2Ab
81.8Abc
84.8Ab
98.0Aa
98.3Aa
95.8Aa
93.9Aa
96.8Aa
89.9Bab
72.1Bc
50.4Be
59.0Bd
46.9Be
32.4Bg
32.5Bfg
46.0Bef
46.8Be
19.0Bh
94.5Aab
98.0Aab
97.6Aab
98.9Aa
99.0Aa
98.8Aa
98.8Aa
98.1Aab
99.2Aa
97.8Aab
95.0Aab
93.1Aab
90.8Ab
96.9Aab
95.9Aab
95.5Aab
87.9Ab
98.8Aa
99.2Aa
96.1Aa
95.0Aab
97.4Aa
92.8Aab
74.4Bc
56.1Bd
73.8Bc
53.6Bd
54.2Bd
42.3Be
53.9Bd
60.0Bd
28.2Bf
a
Paired moisture treatment means within collection date and count day followed by the same superscript do not
differ significantly (P > 0.05). Collection date means within a moisture treatment and count day followed by a similar
subscript do not differ significantly (P > 0.05).
an average of 45 percent by day 21 and 55 percent by day 28
for seed collected from late September to January.
Timing of seed dissemination is determined by seed maturation and factors such as wind, rain, and disturbance by
birds, insects, and mammals. These results suggest that 26
to 72 percent of the seeds that were not disseminated until
after mid-September may enter secondary dormancy. These
seeds that enter secondary dormancy may enter the seed
bank and germinate the following autumn or later. Allowing
this large percentage of seeds to overwinter will potentially
enhance proliferation and perpetuation of Japanese brome
on Northern Great Plains rangelands.
Water Stress ___________________
Seed germination is controlled by the availability of soil
water at an adequate temperature. Pyke and Novak (1994)
found that germination of downy brome occurred over a
normal range of soil water contents, and germination was
not inhibited by soil water potentials of –1.5 MPa. Downy
brome seed tend to respond to intermittent moisture with
minimal delay in germination rate (Allen and others 1994).
Time to germination of downy brome seed may increase as
soil water potential becomes more negative (Evans 1961;
Evans and Young 1972). Our objective was to determine the
impact of water stress on seed germination of Japanese and
downy brome. Alternately, were differences between species
in germination pattern sufficient to explain the dominance
USDA Forest Service Proceedings RMRS-P-31. 2004
of downy brome on drier sites in years following relative dry
autumns?
Methods
Japanese and downy brome seed were collected in July
1992 from native rangeland near Miles City, MT. Seeds
were initially stored in paper sacks in an unheated warehouse until July 1993 when they were brought into the
laboratory and stored at about 21 ∞C. Seeds were germinated in 9-cm petri dishes containing 30 ml of the appropriate solution. Osmotic solutions were prepared with 20,000
MW polyethylene glycol and distilled water. In Trial 1
osmotic potentials ranged from –0.05 to –1.3 MPa, and in
Trial 2 osmotic potentials ranged from –0.03 to –0.99 MPa.
Solution osmotic potentials were measured at the beginning and end of the study with a vapor pressure osmometer.
Each dish contained one piece of a thick medium-speed
filter paper with high retention supported on a polyurethane foam disc (Haferkamp and others 1994a). The filter
paper was continually moistened via a cotton wick inserted
through the center of the foam disc. Once filter paper was
saturated, 100 seeds of each collection were placed on the
filter paper, and covered dishes were placed individually
into a polyethylene bag. Seeds were germinated for 28 days
at alternating 8/23 ∞C with illumination during the 12
hours at 23 ∞C. Dishes were arranged in a randomized
complete-block design with eight replications. Replications
were run over time, with two replications placed in the
117
Haferkamp and MacNeil
Annual Brome Seed Germination in the Northern Great Plains: An Update
25
20
Days
germinator at 2-day intervals in Trial 1 and 7-day intervals
in Trial 2, until all eight replications were being incubated.
Germination was counted every 2 days in Trial 1 and every
7 days in Trial 2. Seeds were considered germinated when
coleoptiles and radicals were 5 mm long.
In Trial 1, a sigmoidal time-dependent response function
was fit to data from each dish. Resulting parameters were
subject to analysis of variance to determine osmotic potential effects on germination patterns. In Trial 2, the sampling
interval precluded fitting a sigmoid-shaped curve to the
data. Thus, the counts of seeds germinated at each sampling
point were subject to analysis of variance directly.
15
10
5
0
-0.63
-0.23
Results and Discussion
-0.12 -0.09 -0.05
Osmotic potential (MPa)
Figure 3—Effect of osmotic potential on days to 50percent germination of Japanese brome seed. Least
square means + standard error at P = 0.05 are
presented.
MPa
110
-0.03
100
Germination (percent)
In Trial 1 no Japanese brome seed germinated in osmotic
potentials ranging from –0.9 to –1.3 MPa (fig. 2). Maximum
seed germination was >95 percent in –0.05 to –0.23 MPa.
Seed germination declined to <35 percent in –0.63 MPa.
Days to 50 percent of maximum germination ranged from
5 to 9 days for –0.05 to –0.23 MPa to 22 days for –0.63 MPa
(fig. 3). In Trial 2 Japanese brome seed germination was
<15 percent in –0.99 MPa (fig. 4), >90 percent in –0.03 to
–0.29 MPa, and <90 percent but >75 percent in –0.5 MPa
and >35 percent in –0.71 MPa. Japanese and downy brome
seed germinated similarly in –0.03 to –0.16 MPa (figs. 4
and 5). Maximum germination of downy brome seed was,
however, up to 5.5 fold greater than germination of Japanese brome seed in –0.29 to –0.99 MPa.
Water stress can both reduce amount and rate of germination of Japanese brome seed. In addition we have observed
years when seedlings that emerged from germinating seeds
during late August perished due to lack of soil water in
September. It appears that soils must be wet for 3 to 5 days
for seeds to germinate. Mack and Pyke (1984) report that a
series of successive days of >1 mm of precipitation during
autumn appears adequate for germination of downy brome.
Thill and others (1979) report that reductions in soil matric
potential from –0.2 to –1.6 MPa markedly reduced the
percentage and rate of germination of downy brome. They
90
-0.09
80
-0.10
70
-0.16
60
50
-0.29
40
-0.50
30
-0.71
20
-0.99
10
0
7
14
21
28
Days
Figure 4—Effect of osmotic potential on percent
germination of Japanese brome seed. Least square
means + standard error at P = 0.05 are presented.
110
MPa
110
-0.03
90
100
80
Germination (percent)
Germination (percent)
100
70
60
50
40
30
20
10
0
-0.63
-0.23
-0.12 -0.09 -0.05
90
-0.09
80
-0.10
70
-0.16
60
50
-0.29
40
-0.50
30
-0.71
20
-0.99
10
0
Osmotic potential (MPa)
7
14
21
28
Days
Figure 2—Effect of osmotic potential on maximum
percentage germination of Japanese brome seed.
Least square means + standard error at P = 0.05 are
presented.
118
Figure 5—Effect of osmotic potential on percent germination of downy brome seed. Least square means +
standard error at P = 0.05 are presented.
USDA Forest Service Proceedings RMRS-P-31. 2004
Annual Brome Seed Germination in the Northern Great Plains: An Update
found that downy brome is adapted to a wide range of
temperatures and soil water conditions, and is limited mostly
by cold soil or warm dry soil. Although seeds of both annual
bromes germinated over a wide range of osmotic potentials
in our study, the fact that downy brome seeds appeared to
germinate with greater stress than Japanese brome seed
helps explain why downy brome dominates on drier sites and
in years preceded by dry autumns in the Northern Great
Plains.
Conclusions ____________________
Annual bromes are clearly adapted to the Northern Great
Plains. The high level of germination exhibited by afterripened seeds suggests a large portion of the disseminated
seeds will germinate with available water during late summer and early autumn. The secondary dormancy state
attained by some seeds will also enhance the species persistence on rangelands, because seedlings emerging in August
and September in any year likely come from two seed crops,
the current and previous years.
Fluctuations in environmental conditions will continue to
cause erratic fluctuations in annual brome populations on
Northern Great Plains rangelands. However, due to the
large seed banks and ability of seeds to germinate in variable
environmental conditions, we think annual bromes will
persist on these rangelands for many years in the future.
Acknowledgments ______________
Authors express appreciation to cooperators Trudy Hentges,
Caralea Leidholt, Debra Palmquist, James Young, Cheryl
Murphy, and Bryon Bennet for assistance in these studies,
and to Mary Ellen French for manuscript assistance.
References _____________________
Allen, P. S.; Debaene-Gill, S. B.; Meyer, S. E. 1994. Regulation of
germination timing in facultatively fall-emerging grasses. In:
Monsen, S. B.; Kitchen, S. G., comps. Proceedings—ecology and
management of annual rangelands; 1992 May 18–21; Boise, ID.
Gen. Tech. Rep. INT-GTR-313. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain Research Station:
215–219.
Baskin, J. M.; Baskin, C. C. 1981. Ecology of germination and
flowering in the weedy winter annual grass Bromus japonicus.
Journal of Range Management. 34: 369–372.
Daubenmire, R. 1970. Steppe vegetation of Washington. Bull. 62.
Pullman: Washington Agricultural Experiment Station. 131 p.
Evans, R. A. 1961. Effects of different densities of downy brome
(Bromus tectorum) on growth and survival of crested wheatgrass
(Agropyron desertorum) in the greenhouse. Weeds. 9: 216–223.
Evans, R. A.; Young, J. A. 1972. Microsite requirements for establishment of annual rangeland weeds. Weed Science. 20: 350–356.
Gartner, F. R.; Butterfield, R. I.; Thompson, W. W.; Roath, L. R.
1986. Prescribed burning of range ecosystems in South Dakota.
In: Hyder, D. N., ed. Proceedings 1st International Rangeland
Congress; 1978 August 14–18; Denver, CO: Society for Range
Management: 687–690.
Haferkamp, M. R.; Grings, E. E.; Heitschmidt, R. K.; MacNeil, M. D.;
Karl, M. G. 2001a. Suppression of annual brome impacts rangeland:
animal responses. Journal of Range Management. 54: 663–668.
Haferkamp, M. R.; Heitschmidt, R. K.; Grings, E. E.; MacNeil, M. D.;
Karl, M. G. 2001b. Suppression of annual bromes impacts
USDA Forest Service Proceedings RMRS-P-31. 2004
Haferkamp and MacNeil
rangeland: vegetation responses. Journal of Range Management. 54: 656–662.
Haferkamp, M. R.; Heitschmidt, R. K.; Karl, M. G. 1997. Influence
of Japanese brome on western wheatgrass yield. Journal of Range
Management. 50: 44–50.
Haferkamp, M. R.; Heitschmidt, R. K.; Karl, M. G. 1998. Clipping
and Japanese brome reduce western wheatgrass standing crop.
Journal of Range Management. 51: 692–698.
Haferkamp, M. R.; Karl, M. G.; MacNeil, M. D. 1994a. Influence of
storage, temperature, and light on germination of Japanese
brome seed. Journal of Range Management. 47: 140–144.
Haferkamp, M. R.; MacNeil, M. D.; Karl, M. G. 1995a. Induction of
secondary dormancy in Japanese brome (Bromus japonicus). In:
th
West, N., ed. Proceedings 5 International Rangeland Congress,
rangelands in a sustainable biosphere; 1995 July 23–28; Salt Lake
City, UT. Denver, CO: Society of Range Management: 199–200.
Haferkamp, M. R.; Palmquist, D.; Young, J. A.; MacNeil, M. D.
1995b. Influence of temperature on germination of Japanese
brome seed. Journal of Range Management. 48: 264–266.
Haferkamp, M. R.; Volesky, J. D.; Borman, M. M.; Heitschmidt, R. K.;
Currie, P. O. 1993. Effects of mechanical treatments and climatic
factors on the productivity of Northern Great Plains rangelands.
Journal of Range Management. 46: 346–350.
Haferkamp, M. R.; Young, J. A.; Grings, E. E.; Karl, M. G.;
Heitschmidt, R. K.; MacNeil, M. D. 1994b. Japanese brome in the
Northern Great Plains. In: Monsen, S. B.; Kitchen, S. G., comps.
Proceedings—ecology and management of annual rangelands;
1992 May 18–21; Boise, ID. Gen. Tech. Rep. INT-GTR-313.
Ogden, UT: U.S. Department of Agriculture, Forest Service,
Intermountain Research Station: 396–401.
Hewlett, D. B.; Johnson, J. R.; Butterfield, R. I.; Mosley, V. K. 1981.
Japanese brome response to atrazine in combination with nitrogen fertilizer in the mixed prairie. Journal of Range Management. 34: 22–25.
Hull, A. C., Jr.; Hansen, W. T., Jr. 1974. Delayed germination of
cheatgrass seed. Journal of Range Management. 27: 366–368.
Hulbert, L. C. 1955. Ecological studies of Bromus tectorum and other
annual brome grasses. Ecological Monographs. 25: 181–213.
Karl, M. G.; Heitschmidt, R. K.; Haferkamp, M. R. 1999. Vegetation
biomass dynamics and patterns of sexual reproduction in northern
mixed-grass prairie. American Midland Naturalist. 141: 227–237.
Mack, R. N. 1981. Invasion of Bromus tectorum L. into Western North
America: an ecological chronicle. Agro-Ecosystems. 7: 145–165.
Mack, R. N.; Pyke, D. A. 1983. The demography of Bromus tectorum:
variation in time and space. Journal of Ecology. 71: 69–93.
Pyke, D. A.; Novak, S. J. 1994. Cheatgrass demography-establishment attributes, recruitment, ecotypes, and genetic variability.
In: Monsen, S. B.; Kitchen, S. G., comps. Proceedings—ecology
and management of annual rangelands; 1992 May 18–21; Boise,
ID. Gen. Tech. Rep. INT-GTR-313. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Intermountain Research Station:
12–21.
Rummell, R. S. 1946. Some effects of competition from cheatgrass
brome on crested wheatgrass and bluestem wheatgrass. Ecology.
27: 159–167.
Thill, D. C.; Schirman, R. D.; Appleby, A. P. 1979. Influence of soil
moisture, temperature, and compaction on the germination and
emergence of downy brome (Bromus tectorum). Weed Science. 27:
625–630.
Vallentine, J. F.; Stevens, A. R. 1994. Use of livestock to control
cheatgrass—a review. In: Monsen, S. B.; Kitchen, S. G., comps.
Proceedings—ecology and management of annual rangelands;
1992 May 18–21; Boise, ID. Gen. Tech. Rep. INT-GTR-313.
Ogden, UT: U.S. Department of Agriculture, Forest Service,
Intermountain Research Station: 202–206.
Whisenant, S. G. 1990. Postfire population dynamics of Bromus
japonicus. American Midland Naturalist. 123: 301–308.
Young, J. A.; Evans, R. A. 1978. Germination requirements as
determinants of species composition of Artemisia rangeland comst
munities. In: Hyder, D. N., ed. Proceedings 1 International
Rangeland Congress; 1978 August 14–18; Denver, CO: Society for
Range Management: 366–369.
119
Fate of Fall-Sown Bitterbrush
Seed at Maybell, Colorado
Robert Hammon
Gary Noller
Abstract: Approximately 50,000 acres of a nearly pure stand of
bitterbrush, Purshia tridentata (Pursh), near Maybell, CO, has
burned in the past 2 decades. Attempts at reclaiming bitterbrush
stands by seeding burned land have been largely ineffective. A
research project funded by the Colorado Division of Wildlife Habitat
Partnership Program was initiated in the fall of 2000 to determine
the causes of seeding failures. Initial observations on seeds planted
in the fall of 1999 indicated that insects such as wireworms and
cutworms may have been responsible in part for seeding failures.
Seeds planted November 2000 had germinated by early April 2001,
with little impact from insect or other predators noted during the
2000/2001 planting and establishment season. More than 90 percent of those seedlings died during the summer from drought. Only
1.25 percent of seeds planted in the fall of 2000 were still alive in
October 2002, following the second year of drought. Seeds treated
with insecticide and fungicide/rodent repellent and untreated seeds
were planted in 10 seed caches on two dates (October 11 and
November 15) in 2001. Samples taken on November 15 and December 19 showed that much of the seed planted on either date had
already germinated. Several fungal pathogens were isolated in the
fall from germinated seed. Rodents, possibly chipmunks, destroyed
much of the November 15 planted seed at one site. Seedling emergence was 2 weeks to 1 month later in 2002 than 2001. Cutworms
killed more than 25 percent of emerged seedlings at one site in April
2002. Most surviving seedlings were killed by drought during the
summer of 2002, and by mid October, only 0.06 percent of fall 2001
planted seeds were still alive.
reseeding resulting in failure. The causes of the failures are
not known, and a research project was funded in 2001 by the
Colorado Division of Wildlife Habitat Partnership Program
to determine the fate of fall-seeded bitterbrush at Maybell.
Maybell Rangeland
Maybell, CO, is located in the extreme northwest corner of
the State in Moffat County. Maybell is located at 40 degrees
N latitude, at an elevation 6,300 ft above sea level. Annual
precipitation is 12 to 15 inches, with about half of the total
moisture falling as snow. The years 2000, 2001, and 2002
have all received lower than normal precipitation (fig. 2).
The average annual temperature is 42 ∞F. Winter temperatures are very cold, with numerous instances of –50 ∞F
recorded. Climactic data for Maybell can be accessed at
http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?comayb. The
rangeland near Maybell is classified as “Sandhills” range,
within Land Resource Area 34. The deep sandy soils are
classified as Cotopaxi loamy sands. They vary from fine
sandy loams in the swales to loamy fine sands on hills and
upland areas.
The research site is dominated by bitterbrush in unburned areas. Other shrubs associated with bitterbrush
are big and silver sagebrush, gray horsebrush, and low and
rubber rabbitbrush. These shrubs are now abundant within
the burned areas. The principal grasses are Indian ricegrass,
needle and thread, sand dropseed, and Sandberg bluegrass. Cheatgrass is a dominant weed in burned areas.
Introduction ____________________
The rangeland west of Maybell, CO (Moffat County), was
at one time the largest continuous stand of bitterbrush,
Purshia tridentata (Pursh), in North America. More than
50,000 acres of this stand has burned since 1980 (fig. 1).
Bitterbrush is a primary winter browse source for many
large game mammals, including a large elk herd that winters in the area. Winter browse has been in short supply
since the fires, and elk are increasingly moving to private
lands, where they are causing considerable damage to haystacks and pastures. The bitterbrush has not regenerated
from seed on the burned lands, with several attempts at
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Robert Hammon, Western Colorado Research Center, 1910 L Road, Fruita,
CO, U.S.A., e-mail: rhammon@lamar.colostate.edu. Gary Noller (Retired),
Upper Colorado Environmental Plant Center, 5538 RBC #4, Meeker, CO
81641, U.S.A.
120
Figure 1—Rangeland near Maybell, CO. The dark
vegetation across the center of the photograph is a
remnant unburned stand of bitterbrush.
USDA Forest Service Proceedings RMRS-P-31. 2004
Fate of Fall-Sown Bitterbrush Seed at Maybell, Colorado
Hammon and Noller
mid-April 2001. A slight amount of insect and rodent feeding
was noted on the seedlings during the spring observations,
but the amount was insignificant. Feeding damage was
observed in both treated and untreated strips. Significant
mortality of emerged seedlings was observed during the
summer of 2001, primarily due to drought. On May 1, 2002,
15 live seedlings were recorded at the Windmill site (3.75
percent survival of planted seeds), and three seedlings were
found at the North site (0.75 percent survival). On October
7, 2002, nine seedlings were alive at the Windmill site and
only one remained alive at the North site. The summer of
2002 was extremely dry (4.25 inches of precipitation through
July), and mortality was attributed to drought.
Precipitation (inches)
2.0
1.5
1.0
0.5
0.0
J F MAM J J A S ON D J F MAM J J A S ON D J F MAM J J
2000
2001
2002
Natural Reproduction of Bitterbrush:
Spring 2001
Month
Figure 2—Precipitation at Maybell, CO, January
2000 to July 2002
Conspicuous forbs are hairy golden aster, buckwheat, lupine, loco, arrowleaf balsamroot, yarrow, death camas, scarlet globemallow, crypthantha, evening primrose, and daisy
fleabane. Percent ground cover in unburned areas is about
25 percent, with ground cover generally less in burned areas.
Fall Planting Studies:
2000 and 2001 __________________
Fall 2000 Study
A simple nonreplicated seeding trial was planted in the
fall of 2000, with the intention of observing seed fate and
seedling behavior to use as a basis for more detailed studies
in 2001. The studies were planted in two wildlife exclosures,
denoted “Windmill” (N 40∞ 27.645'; W 108∞ 11.671'; elevation
6,354 ft) and “North” (N 40∞ 29.260'; W 108∞ 10.943'; elevation 6,314 ft). They are located approximately 5 miles west
of the town of Maybell. Two hundred seeds were planted in
each of two 10-foot-long strips at each site. One strip was
planted with seed treated with imidocloprid (Gaucho 480
FS; 2 oz/cwt) and thiram (Thiram 42-S; 5 oz/cwt), and the
other strip was planted with untreated seed. The Gaucho
480 FS is an insecticide intended to control soil inhabiting
insects such as wireworms and white grubs and aboveground
insect pests such as cutworms. The Thiram 42-S is primarily
used as a fungicide, but also has rodent repellent characteristics when applied at high rates. The rate used in this study
is recommended to repel rodents. The strips were planted on
November 7, 2000. There were approximately 3 inches of
snow on the ground at the time of planting, but the soils were
not yet frozen. There was nearly continuous snow cover
throughout the winter after planting.
The plots were visited on April 4, 2001, at which time
many seedlings were observed. One foot of row from each
strip was dug, and seeds recovered. Germination was calculated to be 79 percent. Seedling emergence was complete by
USDA Forest Service Proceedings RMRS-P-31. 2004
Many natural bitterbrush seedlings were observed in the
area during the spring of 2001. All seedlings observed
appeared to be from rodent-distributed seed caches. The
number of seedlings per cache varied from seven to 30. The
seed caches were most easily found along the sandy edges of
roadways near mature bitterbrush stands. No caches were
observed more than 30 ft from mature bitterbrush stands.
This is probably due to the limited range of the rodents
responsible for burying the seed. There was a slight amount
of insect and rodent feeding damage on natural seedlings,
but it was rare. Significant mortality from drought occurred
during the summers of 2001 and 2002, and by October 2002,
natural caches that had emerged during 2001 were difficult
to find.
Fall 2001 Study
The goal of the fall 2001 study was to evaluate planting
date and seed treatments as factors influencing bitterbrush establishment. The sampling scheme was designed
to determine the causes of seeding failure. The 2001 experiments were conducted in the same wildlife exclosures used
in 2000. The 2001 experiment was a three-factor (planting
date, seed treatment, location) randomized complete block
design with four replications. The seed used in the experiments was produced in 1994 from Maybell Source plants at
the Upper Colorado Environmental Plant Center near
Meeker, CO. All seed was hand planted in 10 seed caches.
1
A planting dibble was used to make a hole 1 to 1 /2 inches
in depth, into which the seed was placed. Plots were seeded
on October 11 and November 15, 2001. The seed treatments
were the same as used in 2000: Gaucho 480 FS (2 oz/cwt)
plus Thiram 42-S (5 oz/cwt) or untreated. Twenty-five
caches of each treatment were planted in each replication.
The October 11 planting date was sampled on November
15, the day the second planting date was seeded. Both
planting dates were sampled on December 19, 2001, April
9, 2002, and May 1, 2002. Five randomly chosen caches
were dug on each sample date to determine seed fate.
Samples were taken to the laboratory where seeds were
removed and inspected for damage. Seed samples from the
December 19 sample were sent to the Jefferson County
Cooperative Extension Plant Diagnostic Clinic, Golden, CO,
in December 2001 for pathogen identification after seed
121
Hammon and Noller
Fate of Fall-Sown Bitterbrush Seed at Maybell, Colorado
rots were observed. More samples from the same sample
date were sent to the North Carolina State University
Plant Pathogen Identification Laboratory, Raleigh, NC, for
molecular characterization of the pathogens. Plots were
evaluated again on June 18, 2002, and October 7, 2002, by
randomly choosing five caches from each treatment per
replication, and counting the live plants.
Results from fall 2001 and spring/summer 2002 sampling
of the fall 2001 planted plots is displayed in tables 1 and 2.
The most surprising aspect of the fall 2001 sampling is that
much of the seed had germinated shortly after planting,
instead of remaining dormant until spring. Seventy-seven
percent of October 11 planted seed had begun the germination process by November 15. Germination of October 11
planted seed was significantly greater at the Windmill
versus the North site, probably a result of drier soil conditions at the latter site. Seed treatments were responsible for
a slight, but significant increase in germination, and reduction in damage of October 11 planted seed on the first sample
date.
When the sites were visited on December 19, 2001, there
were 2 to 4 inches of snow on the ground, and the soil was
frozen. Five caches per treatment were sampled in one replication at each site, and data were not subjected to statistical
analysis. Germination of October 11 planted seed was 92.6
percent on December 19, while that of November 15 planted
seed was 13 percent. Seed planted on November 15 at the
North site was destroyed by rodents between the time of
planting and December 19. Only 20 percent of treated seed
and 4 percent of untreated seed was recovered. Seed that
was cracked and chewed upon was easily found in the
samples. None of the intact seed recovered at the North site
had germinated. Dry conditions, which were responsible in
part for a shallow planting depth at the site, may have made
the seed accessible to rodents. The holes made by the
planting dibble tended to collapse before seed could be
placed into them. This caused seeding depth to be less than
1 inch.
Five random seed caches from each treatment were dug
from each replication on April 9 and again on May 1, 2002.
No seedlings had emerged on April 9, 2002. In 2001, emergence was complete by that date. This difference may have
resulted from differences between the 2 years in soil temperature. Seedlings had emerged on the May 1, 2002, sample
date. Percent emergence, calculated by counting the number
of seedlings divided by the number of seeds planted, was
greatest for the October 11 planted seed at the North site,
and least for the November 15 planted seed at that site.
Differences in emergence at the Windmill site were slight,
although there was significantly greater emergence of treated
seed for the November 15 planting date than the October 11
planting date. On May 1, 6.8 percent of seeds at the Windmill
site and 8.7 percent of seeds at the North site had emerged
and were still alive. There was no significant impact of seed
treatment on emergence, and interactions were not statistically significant for seed treatment x site or planting date
When the plots were visited on June 18, 2002, all emerged
plants at the Windmill site were dead. Only two plants, in a
Table 1—Sample data from fall 2001. The November 15 sample was taken from only October 11 planted seed. The 19 December sample was
nonreplicated, so not subjected to analysis of variance.
Site
Windmill
Planting date
October 11 only
on first sample;
both on second date
North
Seed
treatment
Both
October 11
November 15
October 11 only
on first sample;
both on second date
Both
70.3
NS
70.4b
.0158
October 11
October 11
November 15
November 15
October 11
October 11
November 15
November 15
P-value
Germinated
Dec. 19, 2001
7.1
NS
30.5
46.5
Both
Both
Treated
67.6
76.9
6.6
72.5
82.2a
4.1B
57.5
52
56.0
92.6
13.0
54.7
Untreated
62.8
NS
71.6b
.0373
9.0A
.0727
53.5
51.0
Treated
Untreated
Treated
Untreated
Treated
Untreated
Treated
Untreated
65.0
65.0
86.8
79.9
4.7
7.3
80.0
60.5
77.6
63.2
3.4
10.7
54
78
90
94
60
38
20
4
NS
NS
NS
P-value
Windmill
Windmill
Windmill
Windmill
North
North
North
North
Recovery
Dec. 19, 2001
- - - - - - - - - - - - - - - - - - - - - - - - - - Percent - - - - - - - - - - - - - - - - - - - - - - - - 65.0
83.4a
6.0
79
59.2
Both
P-value
Both
Both
Both
Recovery
Nov. 15, 2001
Sample datesa
Germinated
Damageb
Nov. 15, 2001 Nov. 15, 2001
100
84.6
22.2
29.8
96.7
89.4
0
0
a
Means followed by the same lower case letter are not significantly different at P = 0.05; means followed by the same upper case letter are not significantly different
at P = 0.10.
b
Seed was considered damaged if it appeared abnormal in any way. Most damage was due to chewing by rodents or mold from fungal growth.
122
USDA Forest Service Proceedings RMRS-P-31. 2004
Fate of Fall-Sown Bitterbrush Seed at Maybell, Colorado
Hammon and Noller
Table 2—Sample data from spring 2002.
Sample datesa
April 9, 2002
Site
Planting
date
Seed
treatment
P-value
Recovery
Emergence
October 15
November 15
P-value
80.9a
58.9b
<0.0001
62.0a
40.2b
.0001
65.6a
44.8b
.0001
58.5a
36.8b
<0.0001
18.8
14.3
NS
10.9a
4.8b
.0007
October 15
November 15
October 15
November 15
P-value
81.0a
85.2a
80.7a
32.5b
<0.0001
59.6a
65.0a
64.4a
15.4b
<0.0001
68.5ab
75.8a
62.8b
13.8c
<0.0001
59.0a
58.5a
58.0a
15.0b
.0001
27.7
22.6
9.9
6.1
NS
5.3bc
8.5b
16.5a
1.0c
<0.0001
Untreated
Treated
68.4
71.4
NS
51.0
51.1
NS
56.0
54.4
NS
49.3
46.0
NS
16.9
16.2
NS
7.3
8.4
NS
Untreated
Treated
Untreated
Treated
81.3
85.0
55.5
57.7
NS
62.7
61.9
39.3
40.4
NS
72.5
71.8
39.5
37.0
NS
58.4
59.2
40.3
32.7
NS
20.9AB
29.3A
12.9BC
3.1C
.0835
6.3
7.5
8.3
9.3
NS
October 15
October 15
November 15
November 15
P-value
Untreated
Treated
Untreated
Treated
NS
79.5
82.2
57.3
60.5
NS
62.8
61.2
39.2
41.1
.0860
62.8A
68.5A
49.3B
40.3B
NS
60.0
57.0
38.6
34.9
NS
18.3
19.4
15.5
13.1
NS
10.8
11.0
3.8
5.8
October 15
October 15
November 15
November 15
October 15
October 15
November 15
November 15
P-value
Untreated
Treated
Untreated
Treated
Untreated
Treated
Untreated
Treated
80.5
81.5
82.0
88.5
78.5
83.0
32.5
32.5
NS
68.4a
50.7b
56.9ab
73.0a
57.1ab
71.6a
21.5c
9.2c
.0021
66.5
70.5
78.5
73.0
59.0
66.5
20.0
7.5
NS
63.2a
54.9a
53.5a
63.5a
56.8a
59.2a
23.7b
6.3c
.0309
23.0
32.4
18.9
26.2
13.6
6.2
12.1
0.0
NS
7.0cd
3.5de
5.5cde
11.5bc
14.5ab
18.5a
2.0de
0.0e
.0183
P-value
Windmill
Windmill
North
North
P-value
Windmill
Windmill
Windmill
Windmill
North
North
North
North
Germinated
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Percent - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 83.1a
62.3a
72.1a
58.8a
25.1a
6.9
56.6b
39.9b
38.3b
36.5b
8.0b
8.8
<0.0001
<0.0001
<0.0001
<0.0001
.0027
NS
Windmill
North
Windmill
Windmill
North
North
Recovery
May 1, 2002
Cutworm
Germinated
damage
a
Means followed by the same lower case letter are not significantly different at P = 0.05; means followed by the same upper case letter are not significantly different
at P = 0.10.
single cache, were found in the samples at the North site. The
cause of death of seedlings that had been alive on May 1 was
probably drought. Less than 1/2 inch of precipitation had
fallen between the time seedlings emerged and the sample
date. The plots were evaluated again on October 7, 2002, and
the survival was the same as on June 18. No seedlings were
found at the Windmill site, and only two were found at the
North site.
Insect Predators and Fungal Pathogens
Army cutworm, Euxoa auxiliaris (Grote), killed a significant number of seedlings at both sites; 25.1 percent of
seedlings at the Windmill site and 8.0 percent at the North
site were cut off by May 1, 2002. There were no statistical
differences between planting dates or seed treatments. All
USDA Forest Service Proceedings RMRS-P-31. 2004
recorded cutworm damage occurred between emergence and
May 1.
All soil-inhabiting insects were collected from soil samples
taken in the fall of 2001 and spring of 2002. Several species
of ground beetles (Coleoptera: Carabidae) were found. While
some of these are seed predators, it is not known if they prey
upon bitterbrush seed. White grub larvae (Coleoptera:
Scarabidae: Phyllophaga spp) were common in fall-collected
samples, and adults were found in spring samples. These
root-feeding insects are probably associated with grasses
growing in the area. Seed predation by insects appeared to
be of little significance to seedling establishment. Army
cutworm was the only insect to affect bitterbrush establishment during this study. All specimens collected during the
study are deposited in the Western Colorado Research
Center insect collection at Fruita.
123
Hammon and Noller
100
Germinated
Dormant
Live seed (percent)
Two fungal pathogens were recovered from germinated
seeds during the fall of 2001. They were cultured and
initially identified as Fusarium spp. and Rhizoctonia spp. by
the Jefferson County Plant Disease Diagnostic Laboratory,
Golden, CO. Samples of rotted seed collected on December
19, 2001, were sent to The North Carolina State University
Plant Pathogen Identification Laboratory, Raleigh, NC, for
molecular characterization of the fungus. Two Rhizoctonialike isolates were identified. The most common produced
white mycelia and sclerotia after growth on PDA media. The
other produced brown mycelia, but no sclerotia. Molecular
analysis of these isolates showed that they were actually
Poculum spp., probably unidentified species.
Two Fusarium species were also present, but not common.
One produced a carmine red colony with cottony brown
mycelia. The second isolate produced light pink to yellow
colonies. One isolate was identified as F. redolens, and the
second was determined to be an undescribed Fusarium
species.
Thiram fungicide appeared to have some impact on protecting seedlings after emergence. Seed collected on December 19 was grown out in the greenhouse, and damping off
type symptoms were more common on untreated seed than
treated seed.
Fate of Fall-Sown Bitterbrush Seed at Maybell, Colorado
124
60
40
20
0
June 1995
Sept 1997
March 2002
Test date
Figure 3—Germination tests conducted on 1994
produced seed used in experiments.
Conclusions ____________________
•
Seed Age and Dormancy
The seed used in both the fall 2000 and 2001 planting
studies was produced in 1994 and stored since that time.
Germination tests were conducted in June 1995, September
1997 and March 2002 by the Colorado Seed Laboratory, Fort
Collins, CO. Results from that testing are displayed in figure
3. The percent live seed increased and the percentage of
dormant seed decreased over time. It is interesting to note
that the percent germination recorded in the fall 2001
planting was greater than that recorded in the 2002 seed test
of the same seed lot. The lack of seed dormancy is responsible
for the fall germination of 2001 planting. The 2000 planting
did not experience fall germination because of cold soil
temperatures when the plots were seeded.
80
•
•
•
•
Several factors affected bitterbrush establishment from
seed in this study. Fungal pathogens impacted seed
before emergence. Cutworms killed many plants after
emergence. Drought killed most other plants.
Moisture affected seeding depth and germination, and
lack of moisture killed most seedlings in the first 4
months after emergence.
Once seedlings have survived their first year, they are
much more capable of surviving drought.
Fall germination of seed may be important in pathogenicity of Fusarium- and Rhizoctonia-like organisms.
Gaucho seed treatment was not effective in controlling
insects. Thiram was marginally effective in controlling
pathogens and seed predation by mammals. Other
compounds may be more effective.
USDA Forest Service Proceedings RMRS-P-31. 2004
Comparative Seed Germination
Biology and Seed Propagation of
Eight Intermountain Species of
Indian Paintbrush
Susan E. Meyer
Stephanie L. Carlson
Abstract: In chilling experiments with seeds of eight Castilleja
species, seeds of all species were dormant at harvest and responded
positively to moist chilling. Seeds of species from warmer, drier
habitats had shorter chilling requirements, germinated more quickly
during chilling, and were more likely to lose dormancy in dry storage
than those of species from high elevation habitats with longer
winters. Seed germination patterns within Castilleja were not
affected by the hemiparasitic habit, but instead were similar to
those of species of Penstemon from similar habitats. Seedlings of all
eight species were readily produced in container culture, but
outplanting success was greatly enhanced by potting the seedlings
with a potential host plant for 6 to 8 weeks prior to transplanting
outdoors.
Introduction ____________________
Indian paintbrush (Castilleja: Scrophulariaceae) species
are among the most beautiful, well-known, and beloved
Western native wildflowers, yet they are very seldom seen in
cultivation. Published information on paintbrush propagation and cultural requirements is limited. In this investigation, we attempted to fill this information gap. Our study
was carried out with two objectives in mind, namely to learn
to propagate paintbrushes from seed as well as to understand something about their establishment ecology in the
wild.
The chief obstacle to paintbrush cultivation is often thought
to be their hemiparasitic nature. Paintbrushes are not true
parasites, in the sense that they have chlorophyll and are
capable of making their own food through photosynthesis.
But most, if not all, paintbrushes apparently rely on a root
connection with a host plant as a source of inorganic and
organic nutrients as well as water, especially in more xeric
habitats and during the dry season in mesic habitats (Press
1989). Because of this hemiparasitism, there seems to have
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Susan E. Meyer is a Research Ecologist and Stephanie L. Carlson is a
Biological Technician, U.S. Department of Agriculture, Forest Service, Rocky
Mountain Research Station, Shrub Sciences Laboratory, 735 North 500 East,
Provo, UT 84606, U.S.A., e-mail: smeyer@susna.com
USDA Forest Service Proceedings RMRS-P-31. 2004
been the assumption that paintbrush seeds would be difficult to germinate, perhaps requiring a signal from the host
as in strict parasites such as Striga (Press and Graves 1995).
Thus, reports of even the most straightforward chilling
experiments with these species are lacking in the literature.
In addition, it is not actually known which Intermountain
species are hemiparasites, or whether the relationship with
a host is an obligate one.
In this study, we use a comparative approach in exploring
the relationship between habitat and paintbrush seed germination syndrome. This tactic has been successful for us in
work with many Intermountain genera, including Penstemon, another large genus in the Scrophulariaceae (for example, Meyer and others 1995). The approach involves
making multiple seed collections of related species, and
populations within species, from as wide a range of habitats
as possible, subjecting seeds to a standard array of laboratory germination tests, and relating their responses to habitat at the site of seed origin. This also provides valuable
information for use in propagating these species from seed.
Materials and Methods ___________
Fourteen seed collections representing eight species were
made in the summers of 1997 (six species) and 1998 (C.
scabrida and C. miniata; table 1). The collections were
cleaned by screening and blowing, and seeds were stored in
manila envelopes under laboratory conditions until the
initiation of experiments 2 to 4 months later (depending on
collection date).
The experiment had a completely randomized design and
included six chilling treatments: 0, 4, 8, 12, 16, and 24
weeks of moist chilling at 2 ∞C in the dark. After chilling,
dishes were transferred to a 10/20 ∞C alternating temperature regime (12 hours:12 hours), with cool white fluorescent light during the period at higher temperature. For
each experimental treatment, there were four replications
of 25 seeds per collection. Seeds were placed on the surface
of two moistened germination blotters in plastic 10- by 100mm petri dishes, and blotters were remoistened with water
as necessary during the course of the experiment. Germinated seeds (that is, with radicles visible) were counted and
removed periodically during chilling in the 24-week chilling treatment and weekly during the 4-week postchilling
incubation period for all treatments. At the end of
postchilling incubation, ungerminated seeds were scored
125
Meyer and Carlson
Comparative Seed Germination Biology and Seed Propagation of ...
Table 1—Collection information for 14 seed collections of Castilleja species included in study.
Species
C. exilis
C. exilis
C. linariifolia
C. linariifolia
C. linariifolia
C. chromosa
C. chromosa
C. chromosa
C. scabrida
C. flava
C. miniata
C. applegatei
C. rhexifolia
C. rhexifolia
Population
County
Calf Creek
Utah Lake
North of Boulder
North of Escalante
Boulder Mountain
North of Koosharem
Fishlake Pass
Santaquin Hill
East of Zion
Soldier Summit
Aspen Grove
Bald Mountain Pass
Albion Basin
Mirror Lake
Habitat
Garfield
Utah
Garfield
Garfield
Garfield
Sevier
Wayne
Utah
Kane
Utah
Utah
Summit
Salt Lake
Duchesne
for viability using a cut test (that is, seeds with firm, white
embryos were scored as viable). Mean germination percentages for each collection-treatment combination were
calculated on the basis of total viable seeds. Viability was
generally high (table 1; average 92 percent). Six of the
collections included in this study were also part of a later
study (Mogensen and others 2001) on the effects of chilling
temperature on dormancy loss for species of Castilleja,
Penstemon, and several other genera. These experiments
were initiated in April 2000, approximately 3 years after
harvest for the 1997 paintbrush collections and 2 years
after harvest for the 1998 collections. To examine whether
paintbrush seeds lose dormancy in dry storage, we compared our data for recently harvested seeds with the data
in Mogensen and others (2001) for seeds laboratory stored
for 2 to 3 years. To make the chilling temperatures comparable to our 2 ∞C chilling data, we averaged the data for the
1 and 3 ∞C chilling temperatures.
We analyzed the data for each seed collection in the
chilling experiment with recently harvested seeds using
Viability
Desert riparian
Freshwater marsh
Pinyon-juniper woodland
Pinyon-juniper woodland
Mountain meadow
Sagebrush steppe
Sagebrush steppe
Sagebrush steppe
Pinyon-juniper woodland
Mountain brush
Aspen parkland
Alpine meadow
Subalpine meadow
Alpine meadow
91.5
95.7
90.0
83.7
99.5
87.0
92.7
86.7
88.0
96.5
87.0
96.0
95.8
93.7
one-way fixed effects analysis of variance (ANOVA), with
chilling duration as the independent variable and germination percentage (proportion) as the dependent variable.
Germination proportion was based on the number of viable
seeds in each replicate, and was arcsine transformed to
improve homogeneity of variance prior to analysis.
Results ________________________
Seeds of all eight species of Indian paintbrush were
essentially completely dormant without chilling (table 2).
Species and populations within species varied greatly in
their response, but seeds of all species responded to chilling
to at least some degree. In general, populations from warmer,
low-elevation sites had shorter chilling requirements and
more rapid germination during chilling than populations
and species from higher, colder sites. But there were distinctive differences among species superimposed over this general pattern.
Table 2—Mean germination percentages (expressed as percentage of viable seeds) after chilling at 2 ∞C for 0 to 24 weeks followed by
incubation at 10/20 ∞C for 4 weeks for 14 Utah seed collections belonging to eight Intermountain species of Castilleja.
Chilling duration (weeks)a
Species
C. exilis
C. exilis
C. linariifolia
C. linariifolia
C. linariifolia
C. chromosa
C. chromosa
C. chromosa
C. scabrida
C. flava
C. miniata
C. applegatei
C. rhexifolia
C. rhexifolia
Population
Calf Creek
Utah Lake
North of Boulder
North of Escalante
Boulder Mountain
West of Koosharem
Fishlake Pass
Santaquin Hill
East of Zion
Soldier Summit
Aspen Grove
Bald Mountain Pass
Albion Basin
Mirror Lake
0
0b
0b
3.0b
0c
0d
0d
0c
0c
6.7d
0d
0b
0b
0c
0a
4
8
12
100a
99.0a
100a
90.6b
17.1c
6.5c
5.9c
1.3c
72.1ab
5.7cd
14.7a
1.1b
3.2c
0a
98.8a
100a
100a
93.8b
22.6c
51.5b
48.7b
32.5b
51.0c
10.3c
10.3a
5.3b
11.7b
1.0a
99.0a
100a
100a
98.7a
74.3b
88.3a
83.0a
72.7a
57.2bc
56.3b
11.3a
15.0a
12.4b
2.2a
16
100a
99.0a
100a
100a
100a
81.8a
87.8a
41.0b
56.6bc
66.8b
4.7ab
14.3a
10.0b
1.0a
24
98.9a
100a
99.0a
100a
100a
81.0a
84.9a
66.3a
84.1a
85.9a
13.7a
18.7a
20.6a
6.7a
a
Within a collection (row), means followed by the same letter are not significantly different according to a means separation test using the Student-Newman
Keuls criterion following analysis of variance.
126
USDA Forest Service Proceedings RMRS-P-31. 2004
Comparative Seed Germination Biology and Seed Propagation of ...
Meyer and Carlson
Castilleja exilis
Castilleja linariifolia
C. linariifolia is a common scarlet-flowered summer
species in sagebrush steppe, pinyon-juniper, mountain
brush, and lower montane communities. The seeds from
the three populations included in the study showed marked
differences in germination response to chilling when recently harvested, and these differences were habitat related. The higher elevation Boulder Mountain population
showed a pattern of incremental increase in germination
percentage as a function of chilling duration up to 16 weeks,
while the populations from pinyon-juniper woodland were
largely or completely nondormant after only 4 weeks of
chilling (table 2). The high-elevation collection also germinated much more slowly in the cold (fig. 1). All three lots
germinated completely in the longer chilling regimes, and
there was no evidence of any mechanism for seedbank
carryover.
When the seeds of the high-elevation C. linariifolia collection were subjected to chilling 3 years after harvest, they
showed a dramatic change in chilling response (fig. 2).
Instead of the incremental increase in germination with
chilling duration that was evident for recently harvested
seeds, the 3-year-old seeds germinated to 100 percent after
only 4 weeks of chilling, a response similar to that seen in
recently harvested seeds of lower elevation collections. This
change shows that seeds of this species of Castilleja can lose
dormancy under dry conditions, a process known as dry
after-ripening.
Castilleja chromosa
C. chromosa is the common scarlet-flowered spring
paintbrush species of warm and cool desert shrublands,
sagebrush steppe, and pinyon-juniper woodland in the
Intermountain West. The seeds of all three collections of
C. chromosa in this study were from sagebrush steppe
habitats, and showed similar responses to chilling, although
USDA Forest Service Proceedings RMRS-P-31. 2004
UTAH LAKE
CALF CREEK
80
60
40
20
0
0
4
8
12
16
20
24
28
CASTILLEJA LINARIIFOLIA
100
80
60
40
20
Germination percentage
This scarlet-flowered annual species of wetland habitats
had a short chilling requirement, with full germination of
seeds in postchilling incubation for both populations after
only 4 weeks of chilling (table 2). There was no evidence for
mechanisms, such as the presence of a chilling-nonresponsive fraction, that would promote seedbank carryover across
years. The seeds of this species were very slow to germinate
in chilling, however, particularly those of the northern Utah
Lake lot (fig. 1). The Utah Lake seeds did not begin to
germinate in chilling until 21 weeks, and germinated to less
than 20 percent during chilling, while the southern Calf
Creek lot began germinating in the cold at 18 weeks and
reached about 80 percent by the end of chilling. The chilling
periods required for any germination in the cold are much
longer than those likely to be encountered by these seeds in
their habitat of origin. The remaining seeds of both lots
germinated rapidly and completely after transfer to the
postchilling incubation regime. Germination during chilling
may offer little advantage to this species, which grows in
habitats where water is not limited during the growing
season.
CASTILLEJA EXILIS
100
BOULDER MTN.
N. BOULDER
N. ESCALANTE
0
0
100
4
8
12
16
20
24
28
CASTILLEJA CHROMOSA
80
60
40
20
SANTAQUIN HILL
N. KOOSHAREM
FISHLAKE PASS
0
0
100
4
8
12
16
20
24
28
20
24
28
MID–ELEVATION SPECIES
C. SCABRIDA E. ZION
C. FLAVA SOLDIER SUMMIT
80
60
40
20
0
0
100
4
8
12
16
HIGH–ELEVATION SPECIES
C. RHEXIFOLIA MIRROR LAKE
C. MINIATA ASPEN LAKE
C. APPLEGATEI BALD MOUNTAIN
C. RHEXIFOLIA ALBION BASIN
80
60
40
20
0
0
4
8
12
16
20
24
28
Weeks in chilling (2 °C)
Figure 1—Time courses for germination
during 24 weeks of chilling at 2 ∞C followed by
4 weeks of incubation at 10/20 ∞C for 14
Castilleja seed collections included in the
study . Germination percentages are
expressed as percentage of viable seeds.
The vertical bars represent the point in time
(24 weeks) when the seeds were transferred
from chilling to postchilling incubation.
127
Meyer and Carlson
Comparative Seed Germination Biology and Seed Propagation of ...
Germination percentage
CASTILLEJA LINARIIFOLIA
CASTILLEJA CHROMOSA
CASTILLEJA SCABRIDA
100
100
100
80
80
80
60
60
60
40
40
40
20
20
20
0
0
0
0
4
8
12
16
0
CASTILLEJA FLAVA
4
8
12
0
16
CASTILLEJA APPLEGATEI
100
100
100
80
80
80
60
60
60
40
40
40
20
20
20
0
0
0
4
8
12
16
4
8
12
16
12
16
CASTILLEJA RHEXIFOLIA
0
0
4
8
12
16
0
4
8
Weeks of chilling (2 °C)
RECENTLY HARVESTED
STORED 2-3 YEARS
Figure 2—Changes in chilling responsiveness for six Castilleja seedlots during 2 to 3 years of laboratory
storage. Data for recently harvested seeds are those presented for these lots in table 1, while data for seeds
stored 2 to 3 years are from Mogensen and others (2001).
the more northern Santaquin Hill population was somewhat more dormant overall (table 2). The first substantial
increase in germination took place in response to 8 weeks
of chilling. The germinable fraction for the two southern
populations increased with chilling duration up to 12
weeks, then leveled off between 80 and 90 percent. For the
northern Santaquin Hill population, there was actually a
significant drop in germination percentage between chilling periods of 12 weeks and 16 weeks. The remaining
chilling nonresponsive seeds would have the potential to
carry over across years as a persistent seedbank. Most of
the seeds that became nondormant in chilling germinated
in the cold soon after dormancy loss during the period
from 8 to 12 weeks (fig. 1). This would place germination
in the field in late winter in the sagebrush steppe habitats
where these seeds originated.
Seeds from the Fishlake population tested after an additional 3 years in laboratory storage were largely unchanged
in their chilling response, suggesting that dry after-ripening
plays little role in dormancy regulation, at least for this
population (fig. 2).
Castilleja scabrida
Seeds of this scarlet-flowered Colorado Plateau rock
crevice specialist were less dormant than those of the
closely-related species C. chromosa, germinating to over 70
percent after only 4 weeks of chilling (table 2). They showed
128
decreased germination percentages at intermediate chilling periods of 8 to 16 weeks but germinated to a relatively
high percentage after 24 weeks of chilling. Chilling durations in C. scabrida habitat in the field would never approach 24 weeks, so the information from shorter chilling
durations is probably more ecologically relevant. The presence of a large chilling nonresponsive fraction under these
intermediate chilling regimes is evidence for a mechanism
for seedbank carryover. Germination response under continuous chilling was similar to that for C. chromosa, with
most of the germination taking place between 8 and 12
weeks.
In contrast to those of C. chromosa, the seeds of C. scabrida
definitely afterripened during dry storage for 2 years (fig. 2).
The response to shorter chilling durations was similar for
recently harvested and 2-year-old seeds, but the response to
intermediate chilling durations was greatly increased for 2year-old seeds.
Castilleja flava
Seeds of this golden-flowered mid-elevation mountain
brush species showed incremental increases in germination
percentage as a function of chilling duration up to 24 weeks,
with 86 percent germination in response to 24 weeks of
chilling (table 2). The fact that not all the seeds germinated
even after this prolonged chilling treatment provides some
evidence for seedbank carryover. This species showed an
USDA Forest Service Proceedings RMRS-P-31. 2004
Comparative Seed Germination Biology and Seed Propagation of ...
intermediate germination rate in chilling, with most of the
seeds germinating between 12 and 16 weeks, soon after
dormancy loss (fig. 1).
When the seeds of C. flava were retested after 3 years of
laboratory storage, they showed evidence of after-ripening
in dry storage, with approximately a 20 percent increase in
germination percentage over chilling durations from 4 to 16
weeks (fig. 2).
Castilleja miniata
C. miniata is the common crimson-flowered mid-montane
species of paintbrush throughout the southern Intermountain region and is found in aspen parkland, montane coniferous forest, and mountain meadow communities. The seeds
of the C. miniata seedlot in our experiment were highly
dormant, with a maximum germination of 15 percent and no
incremental increase in response to increasing chilling duration (table 2). It was also relatively slow to germinate in
chilling, with most of the seeds germinating between 14 and
16 weeks (fig. 1).
Castilleja applegatei
C. applegatei is a scarlet-flowered species found in rocky
subalpine and alpine meadows. In our experiment, its seeds
showed an incremental increase in germination percentage
with increasing chilling duration, but the maximum germination was less than 20 percent (table 2). Most of the
germination in the 24-week chilling treatment took place in
the cold between 18 and 21 weeks (fig. 1). When this seedlot
was retested 3 years later, its chilling response was largely
unchanged (fig. 2).
Castilleja rhexifolia
C. rhexifolia is a pale yellow to crimson or purpleflowered species that is closely related to C. miniata and
is found in similar habitats, but is more northerly in its
distribution and ranges up into subalpine and even alpine
communities. The two populations in our study were both
from the upper elevational limits for the species and had
highly dormant seeds. The Albion Basin collection germinated to a maximum of 20 percent, while the Mirror Lake
population had a maximum germination of 7 percent
(table 2). Both seedlots tended to show incremental increases in germination percentage with increasing chilling, but this increase was significant only for the Albion
Basin collection. Neither collection germinated during
chilling; the germination that did occur took place during
postchilling incubation (fig. 1). These results, coupled with
the fact that snow cover commonly lasts 8 to 9 months in
high mountain habitats, suggest that our longest chilling
duration was not long enough to obtain a maximum chilling
response. This statement is also true for the two preceding
species (C. miniata and C. applegatei). Longer chilling
durations could be effective for these species.
Retesting the Albion Basin seed collection after 3 years of
storage gave a chilling response essentially identical to the
one for recently harvested seeds (fig. 2). It is therefore
unlikely that dry after-ripening plays much of a role in
dormancy regulation for this species.
USDA Forest Service Proceedings RMRS-P-31. 2004
Meyer and Carlson
Discussion _____________________
Comparative Germination Ecology
It is clear from our experiments that seed germination
regulation in Intermountain species of Castilleja is not
particularly impacted by the potentially hemiparasitic nature of these plants. The seeds are relatively easy to germinate, responding to moist chilling in a manner essentially
similar to that of many other species of the Intermountain
West. In fact, there are a great many parallels between the
results reported here for Castilleja and those we have
previously published for the related, nonhemiparasitic genus Penstemon (for example, Meyer and others 1995). The
generally obligate requirement for some chilling, even in
after-ripened seeds, is related to the fact that species of both
genera have spring-emerging seedlings.
Variation in laboratory chilling responses can be tied to
mechanisms for timing germination to occur appropriately
in early spring in different habitats. If germination is delayed too long, the seedbed may dry out before the seedlings
have a chance to become established, whereas if germination
occurs too early, long before snowmelt, the newly germinated seedlings are at risk from pathogens that operate at
low temperatures. This is why seeds of populations from
warmer, drier, low-elevation habitats tend to respond to
shorter chilling periods and to germinate more quickly in
chilling than seeds of populations from high-elevation habitats with long winters.
These habitat-correlated differences are seen both within
species with wide ecological amplitude, such as C. linariifolia
in this study, and between species. For example, the three
high-elevation species in this study had very long chilling
requirements, so that even a 6-month chilling period resulted in relatively low germination percentages. This is not
surprising given that chilling conditions under snow cover
often last much longer than 6 months in these habitats. In
contrast, species and populations from southern pinyonjuniper woodland, where long periods of chilling conditions
are unlikely, respond to chilling periods as short as 4 weeks.
Mid-elevation sagebrush steppe and mountain brush species and populations have intermediate chilling requirements that correspond to the chilling periods likely to be
encountered in those habitats.
Most of the species of paintbrush in our study did not
germinate fully even after 24 weeks of chilling. Seedlots of
several of the mid-elevation species contained a sizeable
chilling nonresponsive fraction that would not germinate
even after prolonged chilling, in spite of the fact that the
chilling responsive fraction became nondormant after much
shorter chilling periods. The function of such a chilling
nonresponsive fraction is probably to permit carryover of
seeds across years, even in years when winter conditions
result in longer than average periods of snow cover.
Although most are perennial, many paintbrush species
have populations that turn over fairly rapidly, especially
during cycles of drought. There are often scenarios, particularly at lower elevations, where establishment from a persistent seedbank is necessary for population survival. The
chilling nonresponsive fraction makes this possible. The
mechanism that renders these seeds chilling responsive in
later years may be related to the dry after-ripening we
129
Meyer and Carlson
observed in some of these seedlots. We also have some
preliminary evidence that drying after inadequate chilling
may render the seeds more responsive to chilling the following winter.
The summer annual wetland species C. exilis had a unique
germination response pattern in our study. It had an obligate requirement for chilling that would function to prevent
autumn emergence, but its chilling requirement was short,
only 4 weeks or less. Once the seeds were rendered nondormant in chilling, they remained ungerminated even after
several months in the cold. This is in contrast to the other
species with relatively nondormant seeds, whose seeds germinated in chilling soon after they lost dormancy. C. exilis is
a plant of relatively warm low-elevation habitats where the
length of field chilling would be short, thus necessitating a
short chilling requirement. But unlike low-elevation upland
habitats, the wetland habitat does not pose the problem of a
seedbed that dries rapidly in spring. There would therefore
be no advantage to germinating during chilling for a wetland
species, so it is not unexpected that nondormant C. exilis
seeds do not germinate in the cold.
Seed Propagation
The first steps of seed propagation for species of Castilleja
are essentially similar to those for any native species whose
seeds require moist chilling. The individual seeds within a
lot often have different chilling requirements, and because
the seeds tend to germinate in chilling soon after losing
dormancy, it is difficult to obtain uniform emergence when
ungerminated chilled seeds are planted. By the time most of
the seeds are nondormant, a large fraction has already
germinated, and germinated seeds that remain in the cold
tend to become stringy and difficult to plant successfully. On
the other hand, if the seeds are planted as soon as the first
few begin to germinate, most of the seeds are still dormant
and poor stands are obtained. We can solve this problem on
a small scale by planting the seeds as they germinate in the
cold, while the radicles are still very short, and returning the
remaining seeds to chilling. This results in seedlings of
staggered age, which could be a problem in commercial
production. If seeds are not in short supply, it may be
possible to chill enough so that large batches of seedlings of
each age can be obtained. The germinants are planted
carefully in shallow depressions and covered with a thin (2
to 4 mm) layer of soil. Paintbrush seeds and germinants are
very small, so it is important not to plant them too deeply.
Poor emergence of newly germinated seedlings usually results either from planting too deeply or from allowing the
surface to dry out during emergence. The germinants should
be planted into well-watered potting medium, and the surface should be misted daily, especially if conditions are
sunny, until emergence is complete. At this point it is best to
water less often but more thoroughly, making sure that the
soil is wetted to its full depth each time. Young paintbrush
plants in containers are not as drought hardy as those of
most native species, and it is important to pay careful
attention to the watering. Paintbrushes, perhaps because of
their hemiparasitic nature, tend to have relatively little root
biomass in relation to their shoot biomass. Plants that have
130
Comparative Seed Germination Biology and Seed Propagation of ...
wilted from lack of water sometimes recover by resprouting
from the base, but the resulting plants are not as sturdy as
plants that have been watered consistently.
We prefer a relatively fast-draining potting medium for
Castilleja propagation. Planting containers that produce a
long, narrow root mass, such as root trainers (SpencerLemaire, Edmonton, Alberta) or Ray Leach Conetainers
(Steuwe and Sons, Corvallis, OR), work best for paintbrushes, as they do for most other native herbaceous perennials. The plants can be grown in these containers for 3 to 4
months, depending on growing conditions and container
volume. At this point, when other plants would normally be
hardened for outplanting or potted up individually into
larger containers, our propagation procedure deviates from
the usual pattern.
While we do not have quantitative data to substantiate
our recommendations, we have found through experience
that small paintbrush stock planted out on its own usually
exhibits poor survival, even when planted early in the
spring and even in a garden setting. We have much higher
success rates if we pot the paintbrush seedling up with a
potential host plant and let the two plants grow together for
6 to 8 weeks, then harden them off and outplant them as a
unit. Paintbrush species are not host specific, and we have
had success using shrubs, grasses, and other perennial
herbs as host plants. We particularly like aesthetically
pleasing and ecologically appropriate combinations, such
as planting C. chromosa with various species of sagebrush,
the plants with which it is most often associated in the wild.
Paintbrushes planted out with big sagebrush often live for
many years in gardens and freely self-seed, so that as the
shrub begins to overgrow the paintbrush, more small paintbrushes are produced around its periphery.
In a naturalized garden or restoration project, most species of paintbrush are easily obtained through direct seeding
as long as the seed source used is adapted to the planting
site. If sufficient seed is available, this is clearly the simplest
and least labor-intensive way to establish these plants.
Obviously, autumn planting will be necessary in order to
ensure that the seeds receive the chilling they require in
order to germinate and establish the following spring.
Acknowledgments ______________
We would like to thank Darrin Johnson and Bettina Schultz
for help in obtaining seed collections. Alisa Ramakrishnan
provided invaluable technical assistance.
References _____________________
Meyer, S. E.; Kitchen, S. G; Carlson, S. L. 1995. Seed germination
timing patterns in Intermountain Penstemon. American Journal
of Botany. 82: 377–389.
Mogensen, S. H. A. C.; Allen, P. S.; Meyer, S. E. 2001. Prechill
temperature and duration are important in determining seed
quality for 12 wildflowers. Seed Technology. 23: 145–150.
Press, M. C. 1989. Autotrophy and heterotrophy in root
hemiparasites. Trends in Ecology and Evolution. 4: 258–263.
Press, M. C.; Graves, J. D. 1995. Parasitic plants. London: Chapman
and Hall.
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Biological Soil Crusts
on Seedling Growth and Mineral
Content of Four Semiarid
Herbaceous Plant Species
R. L. Pendleton
B. K. Pendleton
G. L. Howard
S. D. Warren
Abstract: A growing body of evidence indicates that biological soil
crusts of arid and semiarid lands contribute significantly to ecosystem stability by means of soil stabilization, nitrogen fixation, and
improved growth and establishment of vascular plant species. In
this study, we examined growth and mineral content of Bromus
tectorum, Elymus elymoides, Gaillardia pulchella, and Sphaeralcea
munroana grown in soil amended with one of three levels of
biological soil crust material: (1) a low-fertility sand collected near
Moab, UT, (2) sand amended with a 1-cm top layer of excised soil
crust, and (3) crushed-crust material. Plants were harvested after
10 weeks growth, dried, weighed, and leaves were ground for
nutrient analysis. Root architecture was also quantified. Soil crust
additions significantly affected nearly all variables examined. Both
aboveground and belowground biomass were significantly increased
in the presence of crust material. The effect of soil crust additions is
likely due to the increased nitrogen content of the crusts. Nitrogen
tissue content of all four species was greatly enhanced in crusted
soils. All species showed a decline in root/shoot ratios with crust
additions, indicating a shift in plant allocation patterns in response
to improved soil fertility. The proportion of fine feeder roots, as
measured by specific root length, also declined with the addition of
soil crust material. These data support other studies suggesting
that soil crusts have a positive effect on establishment of associated
vascular plant species.
Introduction ____________________
Biological soil crusts are a common surface feature of arid
and semiarid lands throughout the world (West 1990). While
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
R. L. Pendleton and B. K. Pendleton are Research Ecologists, USDA Forest
Service, Rocky Mountain Research Station, 333 Broadway SE, Suite 115,
Albuquerque, NM 87102-3497, U.S.A.; FAX: (505) 724-3688; e-mail:
rpendleton@fs.fed.us. G. L. Howard is Research Biologist, U.S. Army Corps of
Engineers, Construction Engineering Research Laboratory, Champaign, IL,
U.S.A. S. D. Warren is Research Scientist, Center for Environmental Management of Military Lands, Colorado State University, Fort Collins, CO
80523-1500, U.S.A.
USDA Forest Service Proceedings RMRS-P-31. 2004
the ecological role of these crusts varies according to their
composition (Belnap and others 2001; Evans and Johansen
1999), available information indicates that biological crusts
contribute to a variety of ecological functions, including soil
stabilization, nitrogen (N) fixation, nutrient availability,
and vascular plant establishment. We examined the effects
of biological crust additions on the growth, N content, and
root architecture of four plant species: Elymus elymoides,
Gaillardia pulchella, Sphaeralcea munroana, and Bromus
tectorum. Complete results of the study are available in
Pendleton and others (2003).
Methods _______________________
We transplanted 2-week-old seedlings into 6-inch pots
containing one of three soil treatments: (1) a steam-sterilized low-fertility sand collected near Moab, UT, (2) crushed
soil crust material, and (3) 6-inch circles of intact soil crust
placed over the low-fertility sand. We also inoculated plants
with spores of the arbuscular mycorrhizal fungus Glomus
intraradices obtained from Brokow Nursery in Saticoy, CA,
at the time of transplanting. Plants were grown in a greenhouse and bottom watered as needed using a capillary mat
system.
Seedlings were harvested after 10 weeks growth in the
treatment media. Shoots and roots were separated, dried,
and weighed. Ground leaf and stem tissue was analyzed for
N content at the Soil-Plant Analysis Laboratory, Brigham
Young University, Provo, UT, using a micro-Kjeldahl procedure. Dried root tissue was rehydrated and fixed in 70
percent ethanol. Total root length was calculated using a
modified line intersect method (Tennant 1975), and specific
root length was calculated as meters root length per gram
dry root weight. We also confirmed colonization of all plants
by mycorrhizal fungi (Koske and Gemma 1989).
Statistical analyses were performed using SAS version 8.1
for personal computers (SAS Institute 1999–2000). Variables were analyzed separately by species, first using a
MANOVA procedure to determine the significance of treatment effect, followed by univariate analyses using a one-way
ANOVA or GLM procedure.
131
Pendleton, Pendleton, Howard, and Warren
Effects of Biological Soil Crusts on Seedling Growth and Mineral Content of ...
Results and Discussion __________
Soil treatment had a significant effect on growth and N
content of all four plant species (table 1). Plants grown in the
crushed-crust medium were 6 to 8 times the size of plants
growing in sand only. Significant differences also occurred
between plants grown in the crust-over-sand and sand
treatments. Plants grown in crusted soils were larger and
had higher N concentrations in their stem and leaf tissue.
Total uptake of N by plant shoots, calculated as biomass x
concentration, was without exception greatest in the crushedcrust medium, followed by the crust-over-sand treatment.
Other studies have reported increased growth and mineral
content, especially N, of plants grown in crusted soils as
compared to blow sand (Belnap and others 2001).
The relative allocation between root and shoot biomass
was also affected by the addition of crust material (table 1).
Root/shoot ratios of all species decreased with increasing
crust additions, indicating that plants invested proportionately fewer resources in root tissue as nutrient availability
increased (Kachi and Rorison 1989). Values for specific root
length, which measures the relative proportion of fine feeder
roots to larger conducting roots, were also lower in crusted
soils, again indicating less biomass investment in fine feeder
roots for plants growing in higher nutrient soils.
Biological soil crusts contain a large reservoir of bioessential
elements that slowly become available for plant growth. The
high concentration of nutrients relative to the underlying
soil comes from the active fixation of nitrogen and carbon, as
well as the accumulation of small dust particles (Belnap and
others 2001; Harper and Pendleton 1993). Crust material
used in this experiment was high in organic matter, N,
calcium, and magnesium. Next to water, N is the factor most
limiting to plant growth in arid environments (Evans and
Belnap 1999). Mycorrhizal plants in particular may benefit
from the fixation of atmospheric N by soil crusts. Considerable evidence indicates that mycorrhizal fungi play an
important role in N uptake (Hodge and others 2001).
In the Colorado Plateau region, trampling or other compressional disturbance when crusts are dry and brittle
destroys the integrity of the crust, causing a short-lived
flush of nutrients similar to that experienced by plants
growing in the crushed-crust treatment. A rapid increase in
nutrient availability could disrupt community dynamics,
resulting in differential establishment and increased reproductive success of exotic annual species such as Bromus sp.
(McLendon and Redente 1991). Subsequent loss of the crust
community through wind and water erosion reduces the
number of mycorrhizal propagules and other microorganisms associated with the crust layer (Harper and Pendleton
1993). The resulting reduction in soil fertility and stability
of such a degraded ecosystem could have profound and
lasting consequences (Evans and Belnap 1999).
Acknowledgments ______________
This work was supported by grants MIPR 94N107 and
MIPR W2V5AA51113582 from the U.S. Army Corps of
Engineers, Construction Engineering Research Laboratory,
Champaign, IL. Technical assistance was provided by Susan
Garvin, Jeff Ott, David Tarkalson, and Cari York.
Table1—Growth measures and tissue nitrogen content of plants grown in different soil treatments. Letters following means denote significant
treatment differences at P < 0.05 using the Student-Newman-Keuls multiple range test.
Shoot
biomass
Root
biomass
Root/shoot
ratio
---------g--------
Specific root
length
Nitrogen
concentration
Total
shoot N
m/g
percent
mg
Sphaeralcea munroana
Crushed crust
Crust over sand
Sand
P value
3.91a
.51b
.30c
<0.0001
0.86a
.29b
.25b
<0.0001
0.21a
.57b
.84c
<0.0001
7,848a
9,434b
13,054c
<0.0001
2.39a
2.49a
1.99b
.0254
0.925a
.127b
.059c
<0.0001
Gaillardia pulchella
Crushed crust
Crust over sand
Sand
P value
4.66a
.83b
.62c
<0.0001
1.14a
.36b
.31b
<0.0001
.26a
.45b
.51b
<0.0001
9,622a
10,927a
10,846a
.3900
1.30ab
1.45a
1.11b
.0433
.566a
.122b
.067c
<0.0001
Elymus elymoides
Crushed crust
Crust over sand
Sand
P value
1.48a
.30b
.16c
.0073
.26a
.14a
.06a
.1621
.18a
.44b
.46b
.0051
9,731a
16,442b
20,997b
.0137
1.66a
1.65a
1.44a
.3900
.241a
.051b
.023c
.0101
Bromus tectorum
Crushed crust
Crust over sand
Sand
P value
3.19a
.79b
.40c
.0004
.90a
.40b
.22c
.0099
.26a
.54b
.58b
<0.0001
14,496a
24,238b
24,422b
.0051
1.64a
1.09b
.96c
<0.0001
.477a
.083b
.037c
<0.0001
132
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Biological Soil Crusts on Seedling Growth and Mineral Content of ...
Pendleton, Pendleton, Howard, and Warren
References _____________________
nitrogen availability and temperature. Functional Ecology. 3:
549–559.
Koske, R. E.; Gemma, J. N. 1989. A modified procedure for staining
roots to detect mycorrhizae. Mycological Research. 92: 486–488.
McLendon, T.; Redente, E. F. 1991. Nitrogen and phosphorus effects
on secondary succession dynamics on a semiarid sagebrush site.
Ecology. 72: 2016–2024.
Pendleton, R. L.; Pendleton, B. K.; Howard, G. L.; Warren, S. D.
2003. Growth and nutrient content of herbaceous seedlings associated with biological soil crusts. Arid Land Research and Management. 17(3): 271–281.
SAS Institute. 1999–2000. SAS/STAT user’s guide, release 8.1.
Cary, NC: SAS Institute Inc. 1028 p.
Tennant, D. 1975. A test of a modified line intersect method of
estimating root length. Journal of Ecology. 63: 995–1001.
West, N. E. 1990. Structure and function of microphytic soil crusts
in wildlife ecosystems of arid to semiarid regions. Advances in
Ecological Research. 20: 179–223.
Belnap, J.; Kaltenecker, J. H.; Rosentreter, R.; Williams, J.; Leonard,
S.; Eldridge, D. 2001. Biological soil crusts: ecology and management. BLM/ID/ST-01/001+1730, Technical Reference 1730-2.
Denver, CO: National Science and Technology Center, Bureau of
Land Management. 110 p.
Evans, R. D.; Belnap, J. 1999. Long-term consequences of disturbance on nitrogen dynamics in an arid ecosystem. Ecology. 80:
150–160.
Evans, R. D.; Johansen, J. R. 1999. Microbiotic crusts and ecosystem
processes. Critical Reviews in Plant Sciences. 18: 183–225.
Harper, K. T.; Pendleton, R. L. 1993. Cyanobacteria and cyanolichens:
can they enhance availability of essential minerals for higher
plants? Great Basin Naturalist. 53: 59–72.
Hodge, A.; Campbell, C. D.; Fitter, A. H. 2001. An arbuscular
mycorrhizal fungus accelerates decomposition and acquires nitrogen directly from organic material. Nature. 413: 297–299.
Kachi, N.; Rorison, I. H. 1989. Optimal partitioning between root
and shoot in plants with contrasted growth rates in response to
USDA Forest Service Proceedings RMRS-P-31. 2004
133
Seeding Time and Method for
Winterfat in the Semiarid Region
of the Canadian Prairies
Michael P. Schellenberg
Abstract: Seeding recommendations for winterfat (Krascheninnikovia lanata (Pursh) A.D.J. Meeuse & Smit) are to seed in fall. A
3-year establishment study (1997 to 1999) of this semishrub at the
Semiarid Prairie Agricultural Research Centre indicated that yearto-year variation occurs. After-ripened seed from New Mexico was
seeded in fall or in spring. Seeding techniques compared were drill,
broadcast, and broadcast followed by packing in cultivated land.
Seedling counts and dry matter yield data for the establishment and
the second year were obtained. Year-by-year statistical analysis
revealed variation in seeding time benefits for number of seedlings.
Broadcasting resulted in best establishment of winterfat (as found
in the literature), but packing had a negative impact.
Introduction ____________________
Winterfat (Krascheninnikovia lanata) has been recognized as a valuable forage source since the 1890s (Van
Dersal 1938). Anecdotal information indicates that cattle
drives from Texas to the Canadian Prairies followed the
winterfat northward. Also, Canadian ranchers have found
that their best producing winter pastures often had higher
winterfat populations than less productive winter pastures.
Improved economic viability for ranchers in the Northern Great Plains is associated with increased grazing
period with a potential 56 percent increase in economic
efficiency (Saskatchewan Agriculture and Food, Extension
Agrologist, J. Graham, personal communication). This requires a forage source with good nutritional value in the fall
when most forage sources are nutritionally deficient. Crude
protein for many species falls below 7 percent, while winterfat maintains crude protein values above 7 percent
(Clarke and Tisdale 1945), which is sufficient for winter
grazing (Smoliak and Bezeau 1967). As a result, there has
been an increased interest in this shrub due to its good
nutritional composition and late season grazing potential.
Interest, in part, has resulted in increased seed production
as well as development of Northern adapted ecological
varieties. With increased seed availability, the need for
proper planting methods has been realized.
The present recommendation for seeding is fall broadcast
seeding (Romo and others 1997; Stevens and others 1996).
Objectives _____________________
1. Under southwest Saskatchewan climatic conditions, is
fall seeding required?
2. Would placement on the surface of a furrow or packing
after broadcasting improve establishment?
Methods and Materials ___________
A fully randomized factorial design study was seeded 3
years in succession from 1997 to 1999 using seed obtained
from New Mexico (Wind River Seeds) and after-ripened for
a 30-day period. The seeding took place at the Semiarid
Agricultural Research Centre (SPARC), Swift Current (50∞
17’ N, 107∞ 41’ W), Saskatchewan, Canada, on typic
Haploboroll soil (Swinton loam soil [Orthic brown chernozem; Ayres and others 1985]). Each seeding site had four
replicates. The factors examined in this paper were (1) fall
(dormant seeding; late October to November) or spring (late
May to early June) seeding; and (2) seeding method: surface
seeded into a furrow, broadcast, and broadcast followed by
packing.
Data on seedling counts for seeding year and the second
year, as well as dry matter yields for second year, were
collected in September of each year. Meteorological data
were obtained from an Environment Canada weather station. Precipitation and potential evaporation (U.S. Weather
Bureau Class A pan) were totalled for the month. The
precipitation deficit was calculated as total potential evaporation minus total monthly precipitation. Long-term weather
data were obtained from long-term records maintained at
SPARC. Data were analyzed using SAS GLM (SAS Institute
Inc. 1990).
Results and Discussion __________
Meteorological Data
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Michael P. Schellenberg is a Forage Plant Ecologist, Semiarid Prairie
Agricultural Research Center, Agriculture and Agri-Food Canada, Box 1030,
Swift Current, SK, S9H 3X2, Canada. Phone: (306) 778-7247, FAX: (306) 7739123, e-mail: schellenberg@agr.gc.ca
134
Growing season precipitation exceeded the 114-year average for most years of the study (fig. 1). Potential available
moisture also increased from 1998 to 2000 (fig. 2). Precipitation for 1998 and 1999 was lower than 2000 in the winter
months, suggesting less insulative cover (fig. 1). The precipitation for the winter months of 2000 (fig. 1) and mean
monthly temperature (fig. 3) were higher than those of
USDA Forest Service Proceedings RMRS-P-31. 2004
Seeding Time and Method for Winterfat in the Semiarid Region of the Canadian Prairies
Schellenberg
140
Precipitation (mm)
120
100
80
60
40
20
0
Jan
Feb
Mar
Apr
May
July
June
Aug
Sept
Oct
Nov
Dec
Month
1998
2000
1999
114 year mean
Figure 1—Monthly precipitation for 1998 to 2000 and 114-year mean.
previous years. Mean monthly temperatures in the fall of
1997 (fig. 3) were warmer than the following years, resulting
in premature germination and seedling death prior to spring
1998, thus resulting in better 1998 spring seedling establishment (table 1).
Time of Seeding
Spring or fall seeding appears to be year dependant
(table 1) with first-year seedling counts, indicating no
multiple-year benefits for either spring or fall seeding.
With increasing annual growing season precipitation (figs.
1 and 2), fall seeding failed to provide a benefit (table 1). In
the wettest year (2000), no difference between spring or fall
seeding occurred for seedling establishment. Plant counts
for the year following establishment indicate that the effect
of fall seeding continues beyond the seedling year (table 1).
The dry matter yield per plant was better for fall-seeded
plants (table 1). The additional time for growth from earlier
germination is most likely the primary factor.
Method of Seeding
Furrows provide a microenvironment with potentially
increased moisture (Bellotti and Blair 1989) and decreased
wind. However, these potential benefits were negated by
burial of surface-seeded winterfat as indicated by lower
plant counts and dry matter yield (table 2). Broadcasting
and packing failed to result in increased seedling numbers,
although there may have been benefit for dry matter yield
Precipitation deficit (mm)
0
-50
Table 1—Dry matter yields, plant counts for first year and second year
of establishment for time of seeding factor for three seeding
years.
-100
-150
1997–
1998
-200
-250
-300
May
June
July
Aug
Month
2000
1999
1998
Figure 2—Precipitation deficit (Pan “A”) evaporationprecipitation) for 1998 to 2000 growing seasons.
USDA Forest Service Proceedings RMRS-P-31. 2004
Sept
Seeding site
1998–
1999
1999–
2000
Year 1 fall plant count (m–2)
Fall seeding
Spring seeding
2.4
5.8a
2.2a
1.5
1.3
1.4
Year 2 plant count (m–2)
Fall seeding
Spring seeding
nab
nab
1.5
1.2
1.4
1.8
20.2a
9.4
14.3a
8.4
nab
nab
Year 2 dry matter yield (g plant–1)
Fall seeding
Spring seeding
a
b
Statistically significant (P > 0.05).
na = data not available.
135
Seeding Time and Method for Winterfat in the Semiarid Region of the Canadian Prairies
Mean monthly temperature (degrees C)
Schellenberg
15
10
5
0
-5
-10
-15
-20
-25
Jan
Feb
Mar
May
April
June
July
Aug
Sept
Oct
Nov
Dec
Month
1997
1998
1999
2000
Figure 3—Mean monthly minimum temperature.
(table 2). The increased dry matter yield may have resulted
from improved moisture retention or decreased seedling
competition, as a result of lower number of seedlings. Further research is needed to determine exact cause.
Conclusions ____________________
Benefits of fall or spring seeding for winterfat seedling
establishment depended on the prevailing environmental
conditions during the establishment year. The variation in
year could be due to available moisture. The availability of
water, winter temperatures, and available insulative cover
appear to affect the optimum time of seeding. Further
research is required to establish the relationship of water
and temperature.
Table 2—Dry matter yields, plant counts for first year and second year
of establishment for seed method factor for three seeding
years.
1997–
1998
Seeding site
1998–
1999
1999–
2000
Year 1 fall plant count (m–2)
Furrow
Broadcast
Broadcast and packed
2.0a
5.8
4.6b
0.9a
2.3
2.3
0.4a
2.0
1.6b
Year 2 plant count (m–2)
Furrow
Broadcast
Broadcast and packed
nac
nac
nac
0.8a
1.9
2.0
0.6a
2.2
2.0
Year 2 dry matter yield (g plant–1)
Furrow
Broadcast
Broadcast and packed
12.8
14.6
17.1
11.6
10.6
11.8
nac
nac
nac
a
Statistically significant (P > 0.05) difference from broadcast treatments
determined with single degree of freedom contrast.
b
Statistically significant (P > 0.05) difference from broadcast alone treatment
determined with single degree of freedom contrast.
c
na = data not available.
136
Fall seeding did result in greater per plant dry matter
production, although dry matter yields were unaffected by
seeding method. Manipulation of microclimate using furrows did not result in increased seedling establishment.
Broadcasting of winterfat seed resulted in better seedling
establishment compared to seeding in a furrow, but packing
following broadcasting tended to have a negative impact on
seedling numbers.
Acknowledgments ______________
This work was possible with funding support from the
Saskatchewan Agricultural Development Fund and Agriculture and Agri-Food Canada. Technical support from
J. Bolton, B. Hedger, and numerous summer students,
especially C. Neufeld, was greatly appreciated. Poster
development was with the assistance of D. Schott.
References _____________________
Ayres, K. W.; Acton, D. F.; Ellis, J. G. 1985. The soils of Swift Current
map 725 Saskatchewan. Publ. 56. Saskatoon, SK: University of
Saskatchewan, Saskatchewan Institute of Pedology. 22 p.
Bellotti, W. D.; Blair, G. J. 1989. The influence of sowing method on
perennial grass establishment. II. Seedbed microenvironment,
germination and emergence. Australian Journal Agricultural
Research. 40: 313–321.
Clarke, S. E.; Tisdale, E. W. 1945. The chemical composition of
native forage plants of southern Alberta and southwestern
Saskatchewan. Tech. Bull. 46. Ottawa, Ontario: Dominion of
Canada, Department of Agriculture.
Romo, J. T.; Booth, D. T.; Bai, Y.; Hou, J.; Zabek, C. 1997. Seedbank
requirement and cold tolerance of winterfat seedlings: an adapted
forage for the Canadian prairies. Saskatchewan Agricultural
Development Fund Report prepared by: Department of Crop
Science and Plant Ecology, University of Saskatchewan, Saskatoon, SK. 101 p.
SAS Institute, Inc. 1990. SAS users guide: statistics. Cary, NC: SAS
Institute, Inc.
Smoliak, S.; Bezeau, L. M. 1967. Chemical composition and in vitro
digestibility of range forage plants of the Stipa-Bouteloua prairie.
Canadian Journal of Plant Science. 47: 161–167.
USDA Forest Service Proceedings RMRS-P-31. 2004
Seeding Time and Method for Winterfat in the Semiarid Region of the Canadian Prairies
Stevens, R.; Jorgensen, K. R.; Young, S. A.; Monsen, S. B. 1996. Forb
and shrub seed production guide for Utah. Logan: Utah State
University Extension Service. 51 p.
USDA Forest Service Proceedings RMRS-P-31. 2004
Schellenberg
Van Dersal, W. R. 1938. Native woody plants of the United States:
their erosion-control and wildlife value. Washington, DC: U.S.
Government Printing Office. 362 p.
137
New Native Plant Releases From
the USDA-NRCS, Aberdeen, ID,
Plant Materials Center
L. St. John
P. Blaker
Abstract: The Plant Materials Center at Aberdeen, ID, is operated by the United States Department of Agriculture, Natural
Resources Conservation Service. The purpose of the Plant Materials Center is to evaluate and release plant materials for conservation use and to develop and transfer new technology for the
establishment and management of plants. The Center serves
portions of Nevada, Utah, California, Oregon, and Idaho. In 1995,
two selected ecotypes of penstemon were released. The Richfield
Selection of firecracker penstemon has bright red flowers on
upright racemes. The Clearwater Selection of Venus penstemon
has bright lavender flowers on narrow panicles. Native penstemon
species provide soil stabilization, plant diversity, and beautification. ‘Bannock’ thickspike wheatgrass was released in 1995 to be
used as a component of seed mixes for rangeland and pasture
seedings. In 2001, Snake River Plains fourwing saltbush Selected
Class Germplasm and Northern Cold Desert winterfat Selected
Class Germplasm were released. Both of these shrubs are important species on many ecological sites in the Intermountain West
and have potential for use in erosion control, rangeland restoration, livestock and big game browse, and wildlife plantings.
The Plant Materials Center (PMC) at Aberdeen, ID, is
part of the national plant materials program operated by the
United States Department of Agriculture, Natural Resources
Conservation Service. The purpose of the plant materials
program is to develop and transfer new technology for the
establishment, use, and management of plants.
Plant Materials Centers assemble, evaluate, and release
plant materials for conservation use. The Aberdeen PMC
was established in 1939 and has been the primary breeder
and releasing organization for 15 cultivars and 27 alternative releases, and has cooperated in the release of 12 additional cultivars. The Aberdeen PMC serves portions of Nevada, Utah, California, Oregon, and Idaho. The PMC works
cooperatively with private landowners as well as State and
Federal land management agencies. The PMC Farm is
owned by the South Bingham Soil Conservation District and
is leased to the PMC.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
L. St. John is Plant Materials Center Team Leader and P. Blaker is
Administrative Assistant, United States Department of Agriculture, Natural
Resources Conservation Service, Plant Materials Center, P.O. Box 296,
Aberdeen, ID 83210, U.S.A.; e-mail: stjohn@id.usda.gov
138
In 1995, the PMC in cooperation with the Idaho Agricultural Experiment Station released two selected ecotypes of
Penstemon. Except for one minor species, the genus Penstemon does not occur naturally outside of North America.
There are approximately 150 species and most are found in
the Western United States. Penstemons are perennial plants
with opposite leaves, and usually showy, often two-lipped
flowers. The conservation values of penstemon are soil
stabilization, plant diversity, and beautification of many
different sites. The two penstemons were selected from a
collection of 119 penstemons that were evaluated at the
PMC from 1981 to 1985. They were chosen for their natural
beauty, hardiness, and seed production.
The Richfield Selection of Firecracker penstemon (Penstemon eatonii) was collected near Richfield, UT. Its natural
habitat is the sagebrush, juniper, and ponderosa pine zones
at 3,000 to 8,000 ft (914 to 2,438 m) elevation and an annual
precipitation zone from 10 to 16 inches (254 to 406 mm). It
is best adapted to loamy, well-drained soils and can survive
full sunlight, but will not tolerate hot, dry areas. It is not
adapted to areas with poor drainage. Firecracker penstemon
is a perennial, cool-season forb with a fibrous root system,
and the stems are often decumbent or reclining. The leaves
are large and slightly pubescent. The flowers are bright red
on upright racemes 24 to 36 inches (60 to 91 cm) tall.
The Clearwater Selection of Venus penstemon (Penstemon venustus) was collected near the Dworshak Reservoir
on the Clearwater River in northern Idaho. Its natural
habitat is at elevations of 1,000 to 6,000 ft (305 to 1,829 m)
and an annual precipitation zone from 20 to 35+ inches (508
to 889+ mm). Venus penstemon is best adapted to loamy,
well-drained soils and can survive full sunlight on open,
rocky slopes, but does not do well in areas with poor drainage. It is a perennial, cool-season forb, 12 to 24 inches (30 to
60 cm) tall with a strong taproot and woody base. The leaves
are oblong and sharply serrate. The flowers are bright
lavender to purple-violet and appear in one or more narrow
terminal panicles 12 to 20 inches (30 to 51 cm) long.
The PMC and the Idaho Agricultural Experiment Station
cooperatively released ‘Bannock’ thickspike wheatgrass
(Elymus lanceolatus ssp. lanceolatus) in 1995. Bannock is a
composite of six seed collections from Washington, Oregon,
and southeast Idaho. The best performing plants were selected, isolated, and increased to create Bannock. The grass
was developed to be used as a component of a seed mix for
rangeland and pasture seedings. Bannock is especially adapted
to sandy areas with a minimum of 8 inches (200 mm) annual
precipitation. It also provides wildlife cover and nesting.
USDA Forest Service Proceedings RMRS-P-31. 2004
New Native Plant Releases From the USDA-NRCS, Aberdeen, ID, Plant Materials Center
Bannock has been thoroughly tested and compared to
other varieties of thickspike wheatgrass in the Western
United States. It was named in honor of the Bannock Indian
Tribe that inhabited the Great Basin. Bannock is a longlived, leafy, rhizomatous, vigorous, sod-producing, cool-season grass. Thickspike wheatgrass is native to most of the
Northern and Western United States and Southern Canada.
The PMC has been granted Plant Variety Protection (PVP)
for Bannock so that seed can be marketed only as a class of
certified seed. This will help protect and maintain the
characteristics for which it was released.
Snake River Plains fourwing saltbush (Atriplex canescens)
Selected Class Germplasm was released in 2001 by the
Aberdeen PMC, Pullman, WA, PMC, and the Idaho Agricultural Experiment Station. It is a composite of four seed
collections made in 1976 on the Snake River Plains in Power,
Owyhee, and Elmore Counties, Idaho. These were compared
to 79 other collections evaluated at the PMC from 1977 to
1986. The four original collections were selected for their
superior tolerance to cold temperatures.
Fourwing saltbush is one of the most widely distributed
and important native shrubs on rangelands in the Western
United States. Snake River Plains fourwing saltbush is an
erect shrub that can grow to 6 ft (1.8 m) under ideal soil and
moisture conditions. Leaves are simple, alternate, and linear to oblong 1/2 to 2 inches (1.2 to 5.1 cm) long. The species
is mostly dioecious, having separate male and female plants.
Male flowers are red to yellow and form dense spikes at the
end of branches. Female flowers are nondescript in axillary
clusters. The seed is enclosed in a four “winged” membranous utricle.
Snake River Plains fourwing saltbush is potentially
adapted to the northern portion of the Intermountain
USDA Forest Service Proceedings RMRS-P-31. 2004
St. John and Blaker
Western United States where annual precipitation averages
8 to 16 inches (200 to 400 mm). It can be used for erosion
control; rangeland restoration; livestock and big game
browse; and wildlife plantings in dry, moderately saline or
alkaline areas.
The PMC and the Idaho Agricultural Experiment Station
released Northern Cold Desert Winterfat (Krascheninnikovia
lanata) Selected Class Germplasm in 2001. It is a composite
of five seed collections made from 1974 to 1977 and compared
to 40 other collections at the PMC from 1978 to 1986. The five
original seed collections were selected for their superior
tolerance to cold temperatures. Source locations of the original collections selected were Carbon, Emery, Kane, and Washington Counties in Utah, and Rio Blanco County, Colorado.
Winterfat is a widely distributed native shrub ranging
from Saskatchewan and Manitoba, Canada, and northern
Washington to western Nebraska, Colorado, west Texas to
southern California. It is found from near sea level to 10,000
ft (3,000 m) elevation. Northern Cold Desert winterfat is an
erect shrub that can grow to 3 ft (0.9 m). Leaves are simple,
alternate, narrowly linear, flat, with rolled under edges and
densely hairy. Winterfat is monoecious with both male and
female flowers on the same plant. The fruit is a utricle and
the pericarp is thin and covered with fine, white, silky hairs
1
/8 to 1/4 inch (30 to 60 mm) long.
Northern Cold Desert winterfat is potentially adapted to
the northern portion of the Intermountain Western United
States where annual precipitation averages 7 to 16 inches
(175 to 400 mm). It can be used for erosion control; rangeland
restoration; livestock and big game browse; and wildlife
plantings in dry, moderately saline or alkaline areas.
139
Predicting Seedling Emergence
Using Soil Moisture and
Temperature Sensors
Jeffrey R. Taylor
Bruce A. Roundy
Phil S. Allen
Susan E. Meyer
Abstract: Hydrothermal time models are often used to predict seed
germination rates. In this study, soil water potential data from
three resistance-type sensors (Colman cells, Watermark brand
sensors, and Delmhorst gypsum blocks) and from a time-domain
reflectometry (TDR) probe (Campbell Scientific 615) were input into
a hydrothermal time model to predict seedling emergence in a
growth chamber experiment for six desert grass species, including
Brachypodium distachyon, Bromus fasciculatus, Crithopsis
delilianus, and Stipa capensis from the Negev Desert, and naturalized downy brome (Bromus tectorum) and bottlebrush squirreltail
(Elymus elymoides) from the Great Basin Desert. Seeds were sown
in a structureless sandy loam soil that was irrigated approximating
0.58, 1.25, 1.50, and 2.00 times field capacity volumetric water
content and allowed to dry in a growth chamber programmed to
simulate spring temperatures in Provo, UT. Colman cells and
Delmhorst gypsum blocks indicated more rapid soil surface drying
than did Watermark sensors and TDR probes. Inputs provided by
the sensors correctly predicted seedling emergence of these rapidly
germinating grasses when water potential was high, but incorrectly
predicted emergence when rapidly decreasing water potentials
inhibited emergence. These results suggest that more accurate
measurement or prediction of seed zone water potentials may be
necessary before hydrothermal time models can effectively predict
seedling emergence in the field.
Introduction ____________________
Annual and perennial grasses influence arid land plant
communities by affecting biomass production, soil erosion,
plant and herbivore interactions, and fire frequency. In the
Negev Desert, Israel, annual grasses (such as Brachypodium
distachyon, Bromus fasciculatus, Crithopsis delilianus, and
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Jeffrey R. Taylor and Susan E. Meyer are, respectively, Range Technician
and Ecologist at the Shrub Sciences Laboratory, Rocky Mountain Research
Station, Forest Service, U.S. Department of Agriculture, Provo, UT 84606,
U.S.A., e-mail: jrtaylor@fs.fed.us and smeyer@fs.fed.us. Bruce A. Roundy and
Phil S. Allen are, respectively, Professor, Department of Integrative Biology,
and Associate Professor, Department of Plant and Animal Sciences, Brigham
Young University, Provo, UT 84602, U.S.A.
140
Stipa capensis) provide important forage for grazing animals. In the Great Basin Desert, United States of America,
introduced downy brome (Bromus tectorum) can invade and
disrupt native perennial communities (Billings 1994;
D’Antonio and Vitousek 1992).
Successful germination and seedling establishment in
desert ecosystems is often limited by species-specific seed
germination responses to near-surface soil water potentials
and temperatures. Frasier (1987) reported that seedling
survival for five warm-season desert grasses depended upon
the number of seedlings that successfully emerged after
initial wetting, the number of nongerminated and viable
seed that remain, and the time interval between wet and dry
periods. For example, the ability of Lehmann lovegrass
(Eragrostis lehmanniana) to retain a viable seedbank during erratic summer rainfall in the Sonoran Desert grassland
allows it to establish better than some native grasses that
germinate most of their seeds after initial rains and are then
subject to seedling desiccation (Abbott and Roundy 2003).
Seed germination models using soil matric potential (Y)
and temperature data can be used to accurately predict
germination (Allen and others 1999; Bradford 1990; FinchSavage and Phelps 1993; Roman and others 1999).
Gummerson (1986) was the first to combine water potential
and temperature data to model germination (hydrothermal
time). Hydrothermal time has since been used to successfully model germination for a variety of species (Allen and
others 1999; Christensen and others 1997; Finch-Savage
and Phelps 1993; Roman and others 1999).
Electric resistance sensors and time-domain reflectometry (TDR) probes are used to estimate soil moisture conditions (Amer and others 1994; Baker and Allmaras 1990;
Collins 1987; Roundy and others 1997, 2001). Various sensors have been compared to see how they respond to changes
in soil moisture. Topp and Davis (1985) compared TDR and
gravimetric sampling at various depths within the top meter
of soil and found both techniques produced similar results.
Collins (1987) compared gypsum blocks with Colman cells
and found that Colman cells had poor sensitivity in dry soils
as well as large differences among individual sensor resistances. When comparing gypsum blocks and Watermark
sensors, Hanson and others (2000) reported that gypsum
blocks did not indicate reductions in soil water matric
potential until the surrounding media fell below a certain
moisture threshold. In addition, they found that Watermark
USDA Forest Service Proceedings RMRS-P-31. 2004
Predicting Seedling Emergence Using Soil Moisture and Temperature Sensors
Taylor, Roundy, Allen, and Meyer
sensors were more responsive in drier soils. In a comparison
between Colman cells, TDR, and gravimetric sampling,
Amer and others (1994) found that Colman cells tended to
indicate higher soil moisture content and exhibited more
variation, but were more effective at measuring shallow
depths because they measure moisture at points. A more
thorough and simultaneous comparison among available
sensors in soils with widely fluctuating soil water potentials
would better determine their ability to provide meaningful
data to predict seedling emergence.
thermal time assumes that matric potential has no effect (in
other words, it has a value of 0):
Hydrothermal Time
Nondormant seeds germinate after being exposed to suitable temperatures and water potentials for an adequate
period of time (Allen and others 1999; Bradford 1995; FinchSavage and Phelps 1993; Gummerson 1986; Roman and
others 1999). Hydrothermal time can be described by the
equation (Gummerson 1986):
HT = (Y–Yb(g))(T–Tb)t(g)
At the time of germination:
qHT = (Y–Yb(g))(T–Tb)t(g)
HT is the amount of hydrothermal time that has accumulated (expressed as MPa-∞-hours or days). The amount of
hydrothermal time required for germination is qHT (the
amount of HT that satisfies the requirement for radicle
emergence). Y is the water potential of the soil, and Yb(g) is
the minimum or base water potential for (g) fraction of the
seeds to germinate for a given population. T is the temperature of the soil, and Tb is the minimum seed germination
temperature. The time required for g fraction of the seeds to
germinate is t(g).
Hydrothermal time analysis requires that data be probit
transformed (to linearize the typical cumulative time course
for germination of a particular collection). Transformed data
are used to compare germination time differences among the
various fractions of the seed population. Base soil water
matric potentials (Yb) are assumed to vary normally within
the population. The Yb(g) of a given germination fraction
from a population of seeds with a known Yb(50) can be
calculated using a formula:
Yb(g) = Yb(50) + (sYb x prob(g))
where Yb(g) is the minimum or base water potential for g
fraction of the seeds to germinate from a given population,
sYb is the standard deviation of the base water potential, and
prob(g) is the probit conversion of fraction g. Although the
model can algebraically calculate Yb(g) for any fraction,
fractions beyond 2 standard deviations of the mean are less
likely to be normally distributed.
Thermal Time
We adapted the thermal time (TT) equation from the
hydrothermal time (HT) equation. In seed germination studies, TT describes the influence of temperature on seed germination rates. It predicts that radicle emergence will occur
more rapidly as germination temperature increases until
a maximum temperature threshold is reached. Because
USDA Forest Service Proceedings RMRS-P-31. 2004
TT = (0–Yb(g))(T–Tb)t(g)
which can be simplified to:
TT = (–Yb(g))(T–Tb)t(g)
The TT equation and the HT equation incorporate the
same parameters with the exception of soil water potential.
This can be useful for germination modeling when there is
abundant soil moisture but an accurate Y measurement is
unavailable.
Methods _______________________
We had three objectives in this study. First, compare the
drying response of four different soil moisture sensors near
the soil surface with different irrigation treatments. The
sensors we used were TDR probes (Campbell Scientific, Inc.,
Logan, UT), Watermark brand sensors (Irrometer Co., Inc.,
Riverside, CA), Delmhorst gypsum blocks (Delmhorst Instrument Co., Towaco, NJ), and Colman cells (Soiltest Inc.,
Lake Bluff, IL). Second, use soil moisture data from the
different sensors as driving variables in a hydrothermal time
model for 11 grass seed collections from the Negev Desert,
Israel, and the Great Basin, United States of America. Third,
compare predicted and observed emergence for these 11
grass seed collections.
Soil Moisture Release Curve
We used a structureless, nonsaline (electrical conductivity 0.4 mmhos cm–1), sandy loam (74.8 percent sand, 15.2
percent silt, and 8.5 percent clay) for this experiment. Water
content of six soil samples was measured at 0.01, 0.033, 0.1,
0.3, 0.7, and 1.5 MPa using pressure plate Soil Moisture
Equipment Co. extractors. The relationship between volumetric water content and matric tension was modeled using
log-linear regression (double-log) focusing on three data
points in the crucial area:
Y = e –14.3195/ x 4.74341
where Y is soil water matric tension (MPa) and x is volumetric water content (fig. 1).
Sensors
We buried three resistance-type sensors and one TDR
probe in plastic trays. Colman resistance cells were individually calibrated for drying soil water content (Roundy
and others 1997) to reduce the effects of sensor variability,
while standard calibration curves were used for Delmhorst
gypsum blocks (CSI 1983), Watermark sensors (CSI 1996a),
and TDR probes (CSI 1996b). Resistance cells or blocks were
all 2 cm in height or diameter and were buried at a depth of
1 cm (zone of measurement 1 to 3 cm). The TDR probe had
two 3.2 mm diameter rods 30 cm long and spaced 3.2 cm
apart. Rods were buried at 3.5 cm with the effective measurement environment being 2.5 cm above and below the
rods (zone of measurement = 0.5 to 5.5 cm). Soil temperatures
were measured at the same depth as resistance sensors using
141
Taylor, Roundy, Allen, and Meyer
Predicting Seedling Emergence Using Soil Moisture and Temperature Sensors
Matric tension (MPa)
1.6
1.2
y = e -14.3195/x 4.74341
0.8
placed in incubation chambers programmed at 15, 20, and
25 ∞C. We considered a seed to have germinated when the
radicle exceeded 2 mm. Time required for coleoptile emergence after germination was estimated by regressing the
time required for coleoptiles to exceed 3 mm on the temperature of incubation (fig. 2). We estimated time for emergence
after germination using the average thermocouple temperature reading within a tray during a corresponding irrigation
and drying cycle.
0.4
Hydrothermal Time and Thermal Time
Parameters
0
0
0.05
0.1
0.15
Volumetric water content (vol/vol)
Figure 1—Moisture release curve for the sandy
loam soil used in study. Data points indicate observed
values; line is from equation.
thermocouples. Four resistance sensors of each type and one
TDR probe were buried in each of eight trays (four blocks
each receiving one of two irrigation treatments). Trays
measured 60 cm long by 43 cm wide by 15 cm deep. We drilled
holes in the bottoms of each tray to allow drainage. Sensors
were read every minute with CR-10 microloggers (Campbell
Scientific, Inc., Logan, UT), and hourly averages were recorded. We converted soil volumetric water content to soil
water matric potential for the Colman cells and TDR probes
using the moisture release curve (fig. 1).
We assumed the hydrothermal time constant (qHT) and
the base temperature (Tb) were constant for each nondormant seed collection (table 1). Hydrothermal parameters for
fully after-ripened seeds of the collections were developed
using repeated probit regression (Christensen and others
1996; Meyer and others 2000; Taylor and others, unpublished data on file).
Both the hydrothermal time and thermal time models
used an average thermocouple temperature reading from
four locations, each 3 cm deep, for each corresponding tray.
Average soil water matric potential values for each resistance sensor type were calculated similarly while a single
TDR probe provided soil water matric potentials for each
tray.
Growth Chamber Experiment
Fifty seeds from 11 seed collections, including six species,
were sown in each tray prior to irrigation. Collections came
from the Negev Desert of Israel (Brachypodium distachyon
[Brdi, 1996 and 1998], Bromus fasciculatus [Brfa, 1996 and
1998], Crithopsis delilianus [Crde, 1996 and 1998], and
Stipa capensis [Stca 1997]) and the Great Basin of Utah
(Bromus tectorum [Brte 15-Whiterocks, Utah, 1996, 1997–
1999 Mix, Brte 18-Hobble Creek, Utah, 1996], and Elymus
elymoides [Elel, 2000]). We sowed seeds of each collection at
a depth of 3 to 5 mm with the exception of Crithopsis delilianus, which we sowed at a depth of 6 to 8 mm.
Trays were sequentially irrigated and allowed to dry on
two occasions. Trays were initially assigned treatment one
and two and received 1,363 and 2,934 mL of water, respectively (approximately 0.58 field capacity volumetric water
content (fc) or 1.25 fc). Trays were later assigned treatments
three or four and rewetted with 3,521 and 4,695 mL of water,
respectively (approximately 1.50 fc or 2.00 fc). Trays were
arranged in randomized blocks.
We counted seedlings daily and calculated emergence as
a percentage of viable seeds. Seed viability for each collection was determined in preliminary germination tests with
four replications. We placed 25 seeds from each collection in
petri dishes lined with moist blotter paper. Dishes were then
142
Figure 2—Days for coleoptile extension to more
than 3 mm at different incubation temperatures.
Coefficient of determination (r-squared) values
ranged between 0.78 and 0.98.
USDA Forest Service Proceedings RMRS-P-31. 2004
Predicting Seedling Emergence Using Soil Moisture and Temperature Sensors
Taylor, Roundy, Allen, and Meyer
Table1—Hydrothermal time parameters for each of the 11 seed
collectionsa.
Species
B. distachyon
B. distachyon
B. fasciculatus
B. fasciculatus
B. tectorum-15
B. tectorum-15
B. tectorum-18
C. delileanus
C. delileanus
E. elymoides
S. capensis
a
b
Year
1996
1998
1996
1998
1996
Mixb
1998
1996
1998
2000
1997
qHT
(MPa-o-hrs)
672
864
744
480
744
744
672
1296
384
2500
672
Yb(50)
(MPa)
Tb
(C)
sY b
(MPa)
–0.85
–0.58
–1.22
–0.72
–1.01
–1.06
–0.76
–1.08
–0.57
–1.57
–1.23
2.2
2
0.7
1
0
0
0
4.4
10
0
0
0.328
.372
.326
.315
.337
.337
.377
.382
.276
.277
.334
From Taylor, Meyer, Allen, and Roundy, in preparation.
Mix—includes seeds from 1997, 1998, and 1999.
Results and Discussion __________
All soil moisture sensors behaved consistently during the
irrigation and drying cycles. Colman cells and gypsum
blocks indicated more rapid drying than Watermark and
TDR sensors (fig. 3). Colman cells responded to soil drying
most rapidly. The drying pattern for gypsum blocks was
similar to Colman cells, but readings usually lagged 1.5 to 2
days (fig. 3). Watermark sensors responded to the 0.58 fc and
1.25 fc treatments more slowly than the other sensors, but
exhibited the least amount of variation between individual
sensors within their limited range of sensitivity (0 to –0.2
MPa) (fig. 3). Compared with TDR, Colman cells measured
soil moisture over a smaller soil volume. The depth interval
of measurement of the TDR probes was 0.5 to 5.5 cm, making
them the least sensitive to near-surface drying.
The hydrothermal time model and thermal time models
generally predicted similar seedling emergence from water
potential inputs or thermal inputs for each of the sensors,
but usually predicted emergence sooner than actual emergence occurred (fig. 4). Slower actual than predicted emergence may be due to soil impedance of coleoptiles and by
lower soil matric potentials in the seed zone than where the
sensors were buried. The latter explanation is supported by
the results from drier irrigation treatments (fig. 4—fc x 0.58,
fc x 1.25, and fc x 1.50). In three of the irrigation treatments
(0.58, 1.25, and 1.5 fc), both the hydrothermal time model and
thermal time models incorrectly predicted at least 5 percent
relative emergence for some or all of the seed collections before
any corresponding seedlings emerged. With greater irrigation (fig. 4—fc x 2.00), the model overestimated time to 50
percent relative emergence for some collections, but correctly predicted time of emergence for most of the corresponding collections.
The similarity between hydrothermal time predictions
and thermal time predictions (fig. 4) can be explained by two
interrelated causes: the relationship between soil water
USDA Forest Service Proceedings RMRS-P-31. 2004
Figure 3—Soil matric potential measured at 1 to 3 cm
below the soil surface for different sensors in soil
subjected to four irrigation regimes (field capacity x
0.58, 1.25, 1.5, 2.0) and allowed to air dry. Error bars
indicate standard error, n = 4.
143
Taylor, Roundy, Allen, and Meyer
Predicting Seedling Emergence Using Soil Moisture and Temperature Sensors
Figure 4—Observed and predicted time to coleoptile emergence. The models predicted 50
percent relative emergence when there was an observed emergence and 5 percent relative
emergence when there was no observed emergence using matric potential and thermal
inputs for either a hydrothermal time model (matric potential and thermal inputs) or a thermal
time model (thermal inputs only) for each seed collection and irrigation regime. Error bars
indicate standard error, n = 4.
144
USDA Forest Service Proceedings RMRS-P-31. 2004
Predicting Seedling Emergence Using Soil Moisture and Temperature Sensors
Taylor, Roundy, Allen, and Meyer
matric potentials and the period when seeds were germinating and the nature of the hydrothermal and thermal time
equations. Seeds germinated within the first 1 to 4 days after
the 1.25, 1.50, and 2.00 fc irrigation treatments. During this
period, the differences in soil matric potential measurements among the different sensors were minimal, and matric
potential values were relatively high (fig. 3). As a result, the
differences between (Y–Yb(g)) among each sensor type were
minimal. In addition, the influence of (T–Tb) was approximately 10 times greater than the influence of (Y–Yb(g)) or
(–Yb(g)) for the hydrothermal and thermal time equations,
respectively. This large difference in the relative contribution made by the thermal portion of the germination models
essentially masked small differences in soil water matric
potential as measured by each sensor type.
The ability of a hydrothermal time model to accurately
predict emergence of seeds under slower germination and
drying conditions requires sensors that more accurately
measure water potential in the seed zone. Previous research
has successfully created models using soil property related
parameters to estimate daily soil moisture relationships.
Models developed by Finch-Savage and Phelps (1993) are
based on soil temperature and soil-type relationships, while
Roman and others (2000) used weekly soil moisture averages from TDR sensors to develop near-surface soil moisture
relationships. Currently, soil moisture sensors are unable to
provide direct soil water potential inputs that accurately
predict seedling emergence using a hydrothermal time model.
Energy-based heat and waterflow models should be tested to
see if they can predict near-surface water potential to provide accurate hydrothermal time predictions of seedling
emergence.
Billings, W. D. 1994. Ecological impacts of cheatgrass and resultant
fire on ecosystems in the western Great Basin. In: Monsen, S. B.;
Kitchen, S. G., eds. Ecology and management of annual rangelands symposium, proceedings; 1992 May 18–22; Boise, ID. Gen.
Tech. Rep. INT-313. Ogden, UT: U. S. Department of Agriculture,
Forest Service, Intermountain Research Station: 22–30.
Bradford, K. J. 1990. A water relations analysis of seed germination
rates. Plant Physiology. 94: 840–849.
Bradford, K. J. 1995. Water relations in seed germination. In:
Kiegal, J.; Galili, G., eds. Seed development and germination.
New York: Marcel Dekker: 351–395.
Christensen, M.; Meyer, S. E.; Allen, P. S. 1996. A hydrothermal
time model of seed after-ripening in Bromus tectorum L. Seed
Science Research. 6: 1–9.
Collins, J. E. 1987. Soil moisture regimes of rangelands: using
datapods to record soil moisture. In: International conference on
measurement of soil and plant water status: in commemoration
of the centennial of Utah State: proceedings; 1987 July 6–10;
Logan, UT: Utah State University: 219–225.
CSI. 1983. Model 227 Delmhorst cylindrical soil moisture block
instruction manual. Logan, UT: Campbell Scientific, Inc.
CSI. 1996a. Model 253 and 253-L (Watermark 200) soil moisture
sensor. Logan, UT: Campbell Scientific, Inc.
CSI. 1996b. CS615 water content reflectometry instruction manual.
Logan, UT: Campbell Scientific, Inc.
D’Antonio, C. M.; Vitousek, P. M. 1992. Biological invasions by
exotic grasses, the grass/fire cycle, and global change. Annual
Review of Ecological Systems. 23: 63–87.
Finch-Savage, W. E.; Phelps, K. 1993. Onion (Allium cepa L.)
seedling emergence patterns can be explained by the influence of
soil temperature and water potential on seed germination. Journal of Experimental Botany. 44: 407–414.
Frasier, G. W. 1987. Soil moisture availability effects on seed
germination and germinated seed survival of selected warm
season grasses. In: Frasier, G. W.; Evans, R., eds. Seed and
seedbed ecology of rangeland plants: proceedings; 1987 April
21–23; Tucson, AZ. Tucson, AZ: U.S. Department of Agriculture,
Agricultural Research Service.
Gummerson, R. J. 1986. The effect of constant temperature and
osmotic potentials on the germination of sugar beet. Journal of
Experimental Botany. 37: 729–741.
Hanson, B.; Peters, D.; Orloff, S. 2000. Effectiveness of tensiometers
and electrical resistance sensors varies with soil conditions.
California Agriculture. 54(3): 47–50.
Meyer, S. E.; Debaene-Gill, S. B; Allen, P. S. 2000. Using hydrothermal time concepts to model seed germination response to temperature, dormancy loss, and priming effects in Elymus elymoides.
Seed Science Research. 10: 213–223.
Roman, E. S.; Murphy, S. D.; Swanton, C. J. 2000. Simulation of
Chenopodium album seedling emergence. Weed Science. 48:
217–224.
Roman, E. S.; Thomas, A. G.; Murphy, S. D.; Swanton, C. J. 1999.
Modeling germination and seedling elongation of common
lambsquaters (Chenopodium album). Weed Science. 47: 149–155.
Roundy, B. A.; Abbott, L. B.; Livingston, M. 1997. Surface soil water
loss after summer rainfall in a semidesert grassland. Arid Soil
Research and Rehabilitation. 11: 49–62.
Roundy, B. A.; Hegdar, H.; Watson, C.; Smith, S. E.; Munda, B.;
Pater, M. 2001. Summer establishment of Sonoran Desert
species for revegetation of abandoned farmland using line source
sprinkler irrigation. Arid Land Research and Management. 15:
23–29.
Taylor, J. R.; Meyer, S. E.; Allen, P. S.; Roundy, B. A. [In preparation]. A hydrothermal germination model for germination of five
desert annuals of the Negev Desert.
Topp, G. C.; Davis, J. L. 1985. Measurement of soil water content
using time-domain reflectometry (TDR): a field evaluation. Soil
Science Society of America Journal. 49: 19–24.
Acknowledgments ______________
We would like to thank Bruce Webb and the members of
the Brigham Young University Soil and Plant Analysis
Laboratory for helping us operate the soil pressure plate
extractors, and Dr. Chris Grant, Associate Chair of the BYU
Mathematics Department, for helping us model the relationship between volumetric soil water content and soil water
matric potential.
References _____________________
Abbott, L. B.; Roundy, B. A. 2003. Available water influences field
germination and recruitment of seeded grasses. Journal of Range
Management. 56: 56–4.
Allen, P. S.; Meyer, S. E.; Khan, M. A. 1999. Hydrothermal time as
a tool in comparative germination studies. In: Black, M.; Bradford,
K. J.; Vasquez-Ramos, J., eds. Seed biology: advances and applications. CABI Publishing. 401–410.
Amer, S. A.; Keefer, T. O.; Weltz, M. A. 1994. Soil moisture sensors
for continuous monitoring. Water Resources Bulletin. 30: 69–83.
Baker, J. M.; Allmaras, R. R. 1990. System for automating and
multiplexing soil moisture measurement by time-domain reflectometry. Soil Science Society of America Journal. 54: 1–6.
USDA Forest Service Proceedings RMRS-P-31. 2004
145
Community
Ecology
Penstemon acaulis
147
Illustration on the reverse side is by Kaye H. Thorne. In: Fertig, W.; Refsdal, C.; Whipple, J. 1994. Wyoming rare plant field guide. Cheyenne: Wyoming
Rare Plant Technical Committee. Available at: http://www.npwrc.usgs.gov/resource/distr/others/wyplant/species.htm (Accessed 12/18/03).
Antelope Bitterbrush Flowering,
Survival, and Regeneration
Following Ponderosa Pine Forest
Restoration Treatments
Donald J. Bedunah
Kristi D. Pflug
Thad E. Jones
Abstract: To further understanding of the dynamics of antelope
bitterbrush (Purshia tridentata) recruitment in a ponderosa pine
(Pinus ponderosa) forest in western Montana, we measured bitterbrush seed viability and germination, flower numbers in 1997 and
1998, seed depredation, seedling survival between 1994 and 2002,
and population changes between 1992 and 2002 on four forest
restoration treatments. Treatments included a control, a shelterwood
cut, a shelterwood cut followed by a low consumption burn, and a
shelterwood cut followed by a high consumption burn. Bitterbrush
flower numbers did not differ between treatments, but caged bitterbrush had greater flower numbers (P < 0.001) than their uncaged
pairs, indicating that browsing reduced flower numbers. Seed
depredation by rodents and birds caused a small reduction (P <
0.050) in seed crop in 1997. Bitterbrush seed viability and seedling
survival (percent) were both high; although, the number of surviving bitterbrush seedlings averaged only 15 plants per treatment
between 1994 and 2002. In the control treatment there has been a
continual decline in bitterbrush numbers totaling 26 percent between 1992 and 2002; whereas, in the shelterwood cut and burn
treatments bitterbrush numbers have stabilized since 1994. If
bitterbrush stands are desired as a future part of this landscape, it
seems clear that disturbance will be necessary. Failure to allow for
disturbances that reduce forest overstory and increase mineral soil
coverage may ultimately result in loss of bitterbrush from these
stands.
Introduction ____________________
Antelope bitterbrush (Purshia tridentata) is a wide-ranging western shrub found from New Mexico to British Columbia, and from Montana to California in grasslands and open
pine forests. In addition to its intrinsic value as an understory and grassland plant, bitterbrush is often an important
winter browse species for mule deer (Odocoileus hemionus)
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Donald J. Bedunah is Professor, and Kristi D. Pflug and Thad E. Jones are
Graduate Research Assistants, School of Forestry, University of Montana,
Missoula 59812, U.S.A. Fax: (406) 243-6645; e-mail: bedunah@forestry.
umt.edu
USDA Forest Service Proceedings RMRS-P-31. 2004
and elk (Cervus elaphus). In a study by Austin and Urness
(1983), bitterbrush was often the most heavily used browse
species even when making up only 3 percent of the plant
community. Moose (Alces alces), bighorn sheep (Ovis canadensis), yellow-pine chipmunk (Eutamias amoenus), deermice
(Peromyscus maniculatatus), chipping sparrow (Spizella
passerina), and blue grouse (Dendragapus obscurus) are
among the other species that use bitterbrush for food and
cover (Everett and others 1978; Matlock-Cooley 1993; Nord
1965).
During the last five decades, the lack of natural bitterbrush regeneration has become recognized as an increasing
problem throughout much of its range (Clements and Young
1996; Fraas 1992; Peek and others 1978; Young and others
1997). For example, in a study of bitterbrush in six forest
types in Utah from 1957 to 1980, “Purshia population
densities declined in all forest types over the study period.
Average density loss was 35.9 percent in 20 years. Photographs taken in the initial and the 1978 to 1980 periods
demonstrate that many shrub individuals died and left no
replacements” (Harper and Buchanan 1983).
Bitterbrush stands in western Montana have shown a
similar lack of regeneration (Bunting and others 1985).
Studies of the shrub-grass communities at and around the
Mount Haggin Wildlife Management Area in southwestern
Montana have repeatedly stressed the minimal amount of
successful reproduction occurring in that region (Fraas
1992; Guenther 1989; Matlock-Cooley 1993). Such concerns have also been documented in the ponderosa pine
forests of the Lick Creek Study Area in the Bitterroot
National Forest of western Montana (Ayers 1995; Ayers
and others 1999). Mature bitterbrush is an important
component of the understory throughout much of this area;
however, Ayers (1995) found only eight bitterbrush seedlings in the area during the combined field seasons of 1993
and 1994 raising concerns that seedling recruitment would
not replace bitterbrush plants dying of natural causes or
associated with forest restoration treatments (Bedunah
and others 1999). Therefore, the major objectives of our
study were to determine the cause of the low number of
bitterbrush seedlings observed by Ayers (1995) at the Lick
Creek study site and to monitor long-term changes in
bitterbrush populations following ponderosa pine forest
restoration treatments. To determine possible causes of
the lack of regeneration, we measured (1) bitterbrush
149
Bedunah, Pflug, and Jones
Antelope Bitterbrush Flowering, Survival, and Regeneration Following Ponderosa Pine ...
flower numbers; (2) browsing impacts on flower numbers;
(3) seed depredation by ungulates, rodents or birds, and
insects; (4) differences in viability and germination between seeds collected from the site and seeds obtained from
another local seed source, and (5) seedling survival and
population changes between 1994 and 2002. Results from
this study clarify factors associated with bitterbrush regeneration and population changes in the Lick Creek Study
Area as well as provide information for the development of
management prescriptions.
Study Area _____________________
Our study site is the Lick Creek Study Area of the Bitterroot National Forest located 21 km southwest of Hamilton,
MT. Elevations range from 1,311 to 1,402 m. Mean annual
precipitation is 56 cm, with approximately 50 percent in the
form of snow. Soils are of granitic till parent material and are
shallow to moderately deep, with some poorly drained areas
and clay soils at the lowest elevations (Gruell and others
1982). Most of the habitat types (Pfister and others 1977) are
Douglas-fir (Pseudotsuga menziesii) types. The dominant
overstory before treatments was Douglas-fir and ponderosa
pine. The potential site indexes for these two species average
16 m tall at age 50 (Gruell and others 1982; Pfister and
others 1977). The Lick Creek area is considered an important local winter and spring range for mule deer and elk.
White-tail deer and moose also occur but at much smaller
numbers.
In 1991, the Bitterroot National Forest and Intermountain Forest Science Fire Laboratory initiated a project to
examine the response of a ponderosa pine/Douglas-fir stand
to a combination of prescribed fire and shelterwood cutting
as ecological restoration management tools. Photographs of
the study area in the early 1900s are of an open stand of large
ponderosa pine with little shrub understory. Apparently,
with fire suppression and subsequent logging practices, the
area became dominated by dense stands of small diameter
Douglas-fir. In 1992 personnel from the Intermountain Fire
Sciences Laboratory divided the study site into 12 approximately equal 4-ha units. Each unit was assigned to one of the
following four treatments: (1) a shelterwood cut (no burn),
(2) a shelterwood cut and a high consumption burn (high
consumption burn), (3) a shelterwood cut and a low consumption burn (low consumption burn), and (4) a control.
The shelterwood cut was completed in the fall of 1992. Tree
2
basal area was reduced by 53 percent to 13.1m . Before
application of the shelterwood cut, the location of all bitter2
brush within thirty-six 400-m circular plots established in
the control, shelterwood cut, low-consumption burn, and
high-consumption burn treatments were permanently recorded. Prescribed burns were conducted in May 1993. Burn
conditions and bitterbrush survival 2-years post-treatment
are described in Ayers and others (1999).
Methods _______________________
Flower Production
In May 1997, fifteen 0.04-ha plots were chosen from each
of the forest restoration treatments (five plots/replication)
150
established in 1992 to undergo flower counts and seedling
observations. We selected five plots because this was the
minimum number of plots per replication that still contained live bitterbrush plants following the burning treatments. For those replicated treatments having more than
five plots with live bitterbrush, selection was made by first
eliminating plots with less than five remaining live plants,
and then randomly selecting from the remaining plots.
Bitterbrush flowers were counted for the 1997 and 1998 to
determine if forest management treatments significantly
affected flower production. Bitterbrush seedling survival
and change in bitterbrush numbers between 1998 and 2002
were compared with pretreatment and post-treatment bitterbrush numbers. Flower counts were analyzed both as
average number of flowers per hectare and average number
of flowers per plant, as plant numbers varied across plots.
In the fall of 1994, previous researchers randomly selected
20 plants from each forest management treatment and
paired plants of close proximity and with similar vigor and
biomass, caging one plant from each pair in a 1.5-m-high
wire cage (Ayers 1995). During May 1997, we located the
original caged pairs to determine the difference in flower
numbers between caged (unbrowsed) and uncaged (browsed)
bitterbrush. Those pairs that had one or both members dead
or missing were noted as such, and flower count was obtained only for those pairs, which were still viable during
May of 1997 and 1998.
Seed Depredation
At the beginning of July 1997, we reviewed flower count
data to identify those plants that produced at least 20
flowers. Of these, 15 plants per treatment were randomly
selected for monitoring seed production and seed loss. Three
plants per treatment were randomly selected to be caged
with large square (4- by 7.5-cm opening) hardware cloth to
exclude ungulates, three plants were covered with vexar
netting to exclude all seed predators except insects, and
three plants/treatments were unprotected (control). Seed
count was recorded at the beginning of seed production (late
June or early July), and again 3 weeks later (mid to late
July), to determine the ratio of remaining seed count to
original seed count. This ratio was then used to compare the
amount of seed lost before seedfall among the three different
sets of plants to assess whether ungulates, rodents and
birds, or insects were harvesting or browsing seed before
seedfall. In 1998, bitterbrush seeds had already ripened and
fallen before our second seed count, likely a result of record
high temperatures and extremely low levels of precipitation
in July and August. Therefore, seed depredation measurements were not valid for the 1998 growing season and are not
reported.
Seed Viability and Germination
Seed was collected from bitterbrush plants immediately
outside the study plots during the last 2 weeks of July 1997.
The seed were cleaned by hand, removing debris and the
remnant flower parts, which contain germination inhibitors
(Young and Young 1986). After hulling, black or spotted
seeds were removed, as these are usually signs of insect
infestation or not viable (Giunta and others 1978). Viability
USDA Forest Service Proceedings RMRS-P-31. 2004
Antelope Bitterbrush Flowering, Survival, and Regeneration Following Ponderosa Pine ...
rates of Lick Creek seed were compared with viability of seed
purchased from Bitterroot Restoration, Inc. (BRI). BRI bitterbrush seed was collected during the same season about 8
km south of our study site. The BRI seed had been cleaned
and culled in a similar manner to ours and had produced
healthy seedlings in previous greenhouse trials done by
their seed biologist. Both sets of seeds were placed in dry cold
storage (1 ∞C) until testing in the spring of 1998. Viability of
seed was determined using a tetrazolium tests according to
procedures of Grabe (1970) and Meyer and Monsen (1989).
We used four replications of 30 randomly selected seeds from
each seed source.
Germination tests were conducted using procedures of
Young and Evans (1983), Young and Young (1986), and
Meyer and Monsen (1989). Seeds were stratified in the dark
for 6 weeks at 2 to 5 ∞C. Seeds were initially soaked in a 5percent bleach solution for 2 minutes to kill bacteria or mold.
Each replication of 30 seeds was divided into three equal
subsets so that seeds could be spaced on petri dishes to
prevent individual seeds contacting each other. Germination percent for each petri dish was determined as the ratio
of seeds that germinated to the total number of seeds in that
dish.
We also monitored germination and establishment of
planted seeds in the control treatment by planting three
plots of 10 seed groups in November 1997. The seed groups
were planted at a depth of 2.5 cm under mineral soil in
groups of five seeds to simulate natural germination conditions in rodent caches (Evans and others 1983; MatlockCooley 1993). The artificial caches were unmarked to avoid
attracting rodents (Young and others 1997). Sites were
inspected biweekly in May and early June 1998 to record the
number of germinated seeds. Germination percentages were
computed by the percent of seed groups that germinated out
of the total number of seed groups. Seedling survival for the
growing season was measured as the percentage of seedlings
that were surviving in October out of the total number of
germinated seedlings.
Data Analyses
All statistical analyses were conducted using SPSS software version 8.0 (SPSS 1997). Data were tested for normality and homogeneity of variances. In most cases, despite
attempted transformations, most of the data did not meet
assumptions of analysis of variance and necessitated the
use of nonparametric tests. Number of flowers per plant
and number of flowers per hectare between forest restoration treatments and years were compared using the KruskalWallis test (Ott 1993) and the post-hoc comparisons of
Tamhane, Dunnett and Games-Howell (Day and Quinn
Bedunah, Pflug, and Jones
1989). Because these tests would not allow a block by year,
we also used a Kruskall-Wallis test for each year’s data
alone (flower counts), both with and without outliers.
Flower number per plant was compared with all bitterbrush seedlings (any seedling documented between 1993
and 1998) and caged plants excluded from analyses. Flower
count data between caged and uncaged plants and seed
depredation treatments were tested with an approximate
t-test (Day and Quinn 1989). Seed depredation involved
comparing three populations using a two-sample t-test; we
compared each one to the others by pairs (that is, big cage
versus net; big cage versus no cage; no cage versus net).
Comparisons of viability and germination rates between
seed sources were conducted using an approximate t-test
and the nonparametric Wilcoxon Rank Sum/Mann-Whitney
test (Day and Quinn 1989; Norusis 1997) because there
were too few data points to clearly ascertain normality.
Survival of bitterbrush between treatments and years was
tested using a two-way analysis of variance with treatments and years as factors.
Results and Discussion __________
Flower Numbers
Average flower count per hectare was not statistically
different between restoration treatments (P > 0.800) or
years (approximate t-test P = 0.255) (table 1). Average
flower numbers per plant also did not differ between treatments (P > 0.550), but did differ between years (P = 0.074)
(fig. 1). In 1997 and 1998, 57 and 44 percent of bitterbrush
plants produced no flowers, but plants with several hundred to over 2,000 flowers were also found revealing high
plant and plot variability (table 2). We believe the difference in flower production per plant between years was due
to variations in climate, but was also affected by browsing
pressure.
Flower numbers for caged bitterbrush compared to
uncaged pairs were 3.4 and 6 times greater (P < 0.001) in
1997 and 1998, respectively. Caged plants had greater (P <
0.05) flower numbers than the uncaged plants in all treatments except for the control (fig. 2). The reduced flower
numbers on uncaged plants revealed that browsing at the
Lick Creek Study Area has a significant impact on flower
production. We observed many plants during the late
spring and early summer of 1997 that had bark slippage
(stem segments stripped of cambium) on shoots 1 and 2
years old. Two-year-old leaders are heavily involved in
flower production (Shaw and Monsen 1983). Even caged
plants often had considerable twig breakage where ungulates had pushed their heads as far as possible through the
Table 1—Bitterbrush (Purshia tridentata) mean flower counts per hectare (95-percent confidence intervals) in 1997 and
1998 for the different forest restoration treatments at the Lick Creek Study Area.
Year
Control
No burn
1997
1998
7,064 + 4,889
27,037 + 19,278
28,764 + 25,125
45,005 + 41,480
USDA Forest Service Proceedings RMRS-P-31. 2004
Low-consumption
burn
5,579 + 4,702
12,668 + 9,645
High-consumption
burn
4,158 + 2,606
12,035 + 7,532
151
Bedunah, Pflug, and Jones
Antelope Bitterbrush Flowering, Survival, and Regeneration Following Ponderosa Pine ...
2,500
250
200
Flower numbers
Mean flowers/plant
1997
1997
1998
150
100
2,000
1998
1,500
1,000
500
50
0
Control
uncaged
Control
caged
Treated
uncaged
Treated
caged
0
Control
No burn
LCB
HCB
Treatment
Figure 1—Mean bitterbrush (Purshia tridentata)
flowers/plants in 1997 and 1998 in the control,
no-burn (shelterwood cut only), low-consumption
burn (LCB), and high-consumption burn (HCB)
treatments at the Lick Creek Study Area.
tops of the cages. The lack of difference in flower numbers
between caged and uncaged pairs for the control is likely
related to low vigor of bitterbrush plants on the control
treatment. Other researchers (Buwai and Trlica 1977;
Guenther and others 1993) also found that browsing impacted bitterbrush stands, causing decreases in vigor and
biomass. However, Tueller and Tower (1979) found a 70percent reduction in bitterbrush forage production after 2
years of caging. Ferguson and Medin (1983) state that old
bitterbrush will reduce leader growth and increase seed
production if not browsed. Most caged plants on our study
site were three to four times as tall as uncaged plants, and
most of them completely filled their cage. These plants had
been caged for four growing seasons before our initial
flower counts, but were showing none of the reduction in
vigor predicted by Tueller and Tower (1979) or Peek and
others (1978) for unbrowsed plants. However, as stated
previously a number of these caged plants displayed evidence of having been browsed where they extended through
the cage at both top and sides. Perhaps this browsing
stimulation was enough to keep the plants from becoming
senescent. In light of the literature, and the fact that these
caged and relatively lightly browsed plants were much
more vigorous than the uncaged plants, we conclude that
browsing is having a significant negative impact on bitterbrush flower production.
Figure 2—Mean number of bitterbrush (Purshia
tridentata) flowers/plant for the control and
restoration treatments combined for uncaged
and caged (protected from browsing) pairs in
1997 and 1998 at the Lick Creek Study Area.
Flower counts of caged bitterbrush were
significantly different (P < 0.001) than uncaged
pairs for all treatments except the control. The
no burn, low-consumption burn, and highconsumption burn were combined into “treated,”
as there were no differences between these
treatments and all were greater than for the
control (P < 0.01).
Seed Depredation Ratios of Caged,
Netted, and Uncaged Plants
Seed depredation was decreased by netting, but increased by caging plants (P < 0.05). Ratios of remaining
seeds to initial seeds were 0.75, 0.60, and 0.45 for the
netted, uncaged, and large-caged plants, respectively. Apparently, nets kept rodents and birds from consuming
seeds, while cages with large openings provided these
animals with a safe, relatively predator-free place to feed.
We observed a number of caged plants with new chipmunk
holes directly under the plant. This bias created by possibly
improving rodent habitat with these large cages makes it
somewhat difficult to assess possible browsing trends on
seed depredation. However, we do not think that ungulate
browsing had much effect on these seed count ratios because ungulate populations at this time year are small, as
most animals have migrated to their summer ranges. Some
sources (Evans and others 1983; Young and Evans 1978)
cite high seed depredation levels by ants. We observed
Table 2—Bitterbrush (percent) with no flowers, 1 to 10 flowers, 11 to 100 flowers, 101 to 500 flowers, and
greater than 500 flowers in 1997 and 1998 for restoration treatments combined at the Lick Creek
Study Area.
Year
1997
1998
152
None
1 to 10
Flower numbers
11 to 100
101 to 500
Greater than 500
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - percent - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 57.0
14.5
16.9
8.7
3.0
43.7
13.3
19.9
16.0
7.1
USDA Forest Service Proceedings RMRS-P-31. 2004
Antelope Bitterbrush Flowering, Survival, and Regeneration Following Ponderosa Pine ...
Laboratory Viability and Germination
Rates
Comparisons of tetrazolium and laboratory germination
test results revealed no difference in percent viability (P >
0.45) or germination percentages (P > 0.55) between the Lick
Creek and Bitterroot Restoration, Inc. seed sources. Viability and germination percentages averaged 91 and 28 percent, and 91 and 33 percent, respectively for the Lick Creek
and BRI seed sources. In theory, tetrazolium and laboratory
germination tests should yield similar percentages (Grabe
1970; Meyer and Monson 1989; Meyer and others 1986).
However, mold is frequently a problem in laboratory stratification and germination, causing some authors to recommend tetrazolium testing only (Grabe 1970; Meyer and
others 1986). Once germination testing began we could not
bleach or otherwise eliminate mold because the chemicals
used would probably also have killed the embryos. This
contamination problem is the reason for the lower germination percentages compared to viability. In our field germination trials, 75 percent of the seeds germinated, and 87
percent of these survived through the first growing season.
Seedling Recruitment and Change in
Bitterbrush Numbers
Total seedling recruitment, measured as seedlings identified pre-1998 and still present in 2002, was 4, 32, 8 and 4
plants for the control, no burn, low-consumption burn, and
high-consumption burn treatments, respectively. Seedling
survival was 78 percent across treatments with no difference (P = 0.69) between treatments. Thus, the number of
bitterbrush seedlings located has been small (and not significantly different between treatments), but survival has
been high. The lack of seedlings is a concern in that the
initial decrease in bitterbrush numbers following treatments averaged 34, 62, and 65 percent for the no burn, lowconsumption burn, and high-consumption burn treatments
(fig. 3). Seedling recruitment in these treatments has not
been able to add significantly to the population losses caused
by the initial treatments, with the no burn, low-consumption
burn, and high-consumption burn treatments averaging
103, 106, and 99 percent of bitterbrush numbers found in
1994. However, bitterbrush in the control treatment has
declined by 26 percent (P < 0.05) since the study began in
1992, showing significant mortality for untreated stands
(fig. 3). The decrease in bitterbrush in the control treatment
appears to be a relatively continuous decline, with 18percent mortality between 1994 to 1998, and an additional
8-percent mortality between 1992 to 2002 (fig. 3). We believe
this mortality is associated with the poor vigor of bitterbrush
in these undisturbed forest stands and illustrates that a “no
treatment” option is not a viable option for maintaining
bitterbrush in the Lick Creek area.
USDA Forest Service Proceedings RMRS-P-31. 2004
120
100
80
Percent
minimal ant activity on these plants during our seed count
measurements. We conclude that since uncaged plants
maintained 60 percent of their seed until seedfall, seed
depredation was probably not a significant problem during
1997.
Bedunah, Pflug, and Jones
60
40
20
0
Control
Change – 1992-1994
No burn
LCB
Change – 1994-1998
HCB
Change – 1994-2002
Figure 3—Change (percent) in bitterbrush (Purshia
tridentata) numbers between 1992 to 1994, 1994 to
1998, and 1994 to 2002 for a control, no burn
(shelterwood cut only), low-consumption burn (LCB),
and high-consumption burn (HCB) at the Lick Creek
Study Area. Pretreatment bitterbrush numbers were
184, 258, 259, and 490 for the control, no burn, lowconsumption burn, and high-consumption burn
treatments, respectively.
We examined historical photos taken of our study area
during the past 90 years (some from Gruell and others 1982,
others from the U.S. Forest Service Region 1 Headquarters,
Missoula, MT) to determine the historical density and size of
bitterbrush on this site. These pictures show mostly small,
low-growing bitterbrush at low density for the areas photographed in 1909. These photographs were taken 14 years
after the last recorded fire, which is within the 3- to 30-year
fire interval for the site (Arno 1976). By the late 1920s,
bitterbrush in the photographs are larger and the stands
more dense. However, as the forest canopy closes no bitterbrush regeneration is visible, and stands appear less vigorous. When trees are removed in subsequent thinning, the
bitterbrush stands appear to regain vigor. Photo interpretation indicates that bitterbrush were present on this site
during the era of high-frequency, low-intensity fires although in smaller form and numbers (for example, see photo
series accompanying USDA Forest Service photos 87357
and 86480). Therefore, the population levels of bitterbrush
now found on the restoration treatments, although lower
than before treatments, may represent more natural (pre1900) levels of bitterbrush.
Summary and Management
Implications ____________________
The primary goal of this study was to understand bitterbrush regeneration and survival following forest restoration
treatments at the Lick Creek Study Area. The restoration
treatments were designed to restore the area to an open
ponderosa pine stand, similar to conditions prior to European settlement. Previous studies have shown high bitterbrush mortality on all treatments except the control and
noted almost no bitterbrush flower production or seedlings
153
Bedunah, Pflug, and Jones
Antelope Bitterbrush Flowering, Survival, and Regeneration Following Ponderosa Pine ...
2 and 3-years post-treatment. We found no significant problems in bitterbrush seed viability, field germination, or
seedling survival. Netting plants significantly reduced seed
depredation; however, depredation of seeds on unprotected
plants was only 40 percent and does not appear to be a major
factor for the low seedling recruitment. Much higher flower
numbers for caged bitterbrush compared to uncaged pairs
shows that browsing by ungulates impacts flower numbers.
It seems likely that more bitterbrush regeneration would
occur if browsing pressure was reduced, especially where
there is not a buildup of undecomposed organic matter and
suitable sites for rodent caching and seed germination.
Several studies have shown that a buildup of undecomposed
organic matter reduces suitable sites for rodent caching and
seed germination (Evans and others 1983; Ferguson and
Medin 1983; Fraas 1992; Matlock-Cooley 1993).
The no burn, low-consumption burn, and high-consumption burn treatments resulted in 34, 62, and 65 percent
bitterbrush mortality, respectively, following treatments
(1992 to 1994). Bitterbrush mortality was found to be significantly related to mechanical damage class and burn severity
(Ayers and others 1999). Since 1994, bitterbrush numbers
have stabilized for these treatments, with the low consumption burn and no burn treatments having small increases in
total plants (although not significantly different); however,
for the control (no treatment) bitterbrush numbers decreased by 26 percent. Therefore, it is apparent that a “no
treatment” is not a viable option for maintaining bitterbrush
in the Lick Creek area. In addition, the potential of a highintensity wildfire is much greater in stands with no treatment as compared to forest restoration treated stands. A
high-intensity wildfire would likely result in extreme mortality of bitterbrush in this area. Land managers may want
to consider a mosaic of different treatments on the landscape
as a simulation of mosaics created by fires in the past. Such
mosaics provide more diversity of habitat and species, both
flora and fauna, and less potential for landscape-scale forest
stand replacement by insects, disease, or fire (Camp and
others 1996). A shelterwood cut with no burning would
result in less bitterbrush plant loss compared to a treatment
with understory burning. However, disturbances mimicking natural events, such as forest stand thinning and lowintensity burns, which maintain open ponderosa pine stands,
should allow for bitterbrush maintenance and further reduce the potential of wildfires. The type of treatment chosen
will depend on the specific management objectives for a
given area, both in terms of bitterbrush stand condition and
in terms of timber and other interests. However, it is apparent that if bitterbrush stands are desired as a future part of
this landscape, some type of disturbance will be necessary to
reduce forest overstory and increase mineral soil coverage
on at least a portion of the landscape. Failure to provide or
allow for these disturbances may ultimately result in longterm loss of bitterbrush from these stands, and for other
species that also depend on disturbance for regeneration and
maintenance.
References _____________________
Arno, S. F. 1976. The historical role of fire on the Bitterroot National
Forest. Res. Pap. INT-187. Ogden, UT: U.S. Department of
154
Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station. 29 p.
Austin, D. D.; Urness, P. J. 1983. Summer use of bitterbrush
rangelands by mule deer. In: Tiedemann, A. R.; Johnson, K. L.,
comps. Proceedings: research and management of bitterbrush
and cliffrose in Western North America. Gen. Tech. Rep. INT152. Ogden, UT; U.S. Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment Station: 203–212.
Ayers, D. M. 1995. Response of bitterbrush and Scouler’s willow to
a shelterwood harvest and prescribed burn in western Montana.
Missoula: University of Montana. 37 p. Thesis.
Ayers, D. M.; Bedunah, D. J.; Harrington, M. G. 1999. Antelope
bitterbrush and Scouler’s willow response to a shelterwood harvest and prescribed burn in western Montana. Western Journal
of Applied Forestry. 14: 137–143.
Bedunah, D. J.; Harrington, M. G.; Ayers, D. M. 1999. Antelope
bitterbrush and Scouler’s willow response to a forest restoration
project. In: McArthur, E. D.; Ostler, W. K.; Wambolot, C. L.,
comps. Proceedings: shrubland ecotones; 1998 August 12–14;
Ephraim, UT. Proc. RMRS-P-11. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Rocky Mountain Research Station:
256–259.
Bunting, S. C.; Neuenschwander, L. F.; Gruell, G. E. 1985. Fire
ecology of antelope bitterbrush in the Northern Rocky Mountains. In: Fire effects on wildlife habitat, symposium proceedings.
Gen. Tech. Rep. INT-186. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station.
Buwai, M.; Trlica, M. J. 1977. Multiple defoliation effects on herbage
yield, vigor and total nonstructural carbohydrates of five range
species. Journal of Range Management. 30: 164–171.
Camp, A. E.; Hessburg, P. F.; Everett, R. L. 1996. Dynamically
incorporating late-successional forest in sustainable landscapes. In: Hardy, C. C.; Arno, S. F., eds. The use of fire in forest
restoration. Gen. Tech. Rep. INT-GTR-341. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain
Research Station: 20–23.
Clements, C. D.; Young, J. A. 1996. Influence of rodent predation on
antelope bitterbrush seedlings. Journal of Range Management.
49: 31–34.
Day, R. W.; Quinn, G. P. 1989. Comparisons of treatments after an
analysis of variances in ecology. Ecological Monographs. 59: 433–463.
Evans, R. A.; Young, J. A.; Cluff, G. J.; McAdoo, J. K. 1983. Dynamics
of antelope bitterbrush seed caches. In: Tiedemann, A. R.; Johnson,
K. L., comps. Proceedings: research and management of bitterbrush and cliffrose in Western North America. Gen. Tech. Rep.
INT- 152. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range Experiment Station:
192–202.
Everett, R. L.; Meeuwig, R. 0.; Stevens, R. 1978. Deer mouse preference for seed of commonly planted species, indigenous weed seed,
and sacrifice foods. Journal of Range Management. 31: 70–73.
Ferguson, R. B.; Medin, D. E. 1983. Long-term changes in an
ungrazed bitterbrush plant community in southwest Idaho. In:
Tiedemann, A. R.; Johnson, K. L., comps. Proceedings: research
and management of bitterbrush and cliffrose in Western North
America. Gen. Tech. Rep. INT- 152. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station: 107–116.
Fraas, W. W. 1992. Bitterbrush growth and reproductive characters
in relation to browsing levels in southwest Montana. Bozeman:
Montana State University. 137 p. Thesis.
Giunta, B. C.; Stevens, R.; Jurgensen, K. R.; Perry, A. 1978.
Antelope bitterbrush. Publ. 78-12. Utah Division of Wildlife
Research. 48 p.
Grabe, D. F., ed. 1970. Tetrazolium testing handbook for agricultural seeds. Contribution 29 to the handbook on seed testing.
Association of Official Seed Analysis. 62 p.
Gruell, G. E.; Schmidt, W. C.; Arno, S. F.; Reich, W. J. 1982. Seventy
years of vegetative change in a managed ponderosa pine forest in
western Montana—implications for resource management. Gen.
Tech. Rep. INT-130. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Forest and Range Experiment
Station. 42 p.
USDA Forest Service Proceedings RMRS-P-31. 2004
Antelope Bitterbrush Flowering, Survival, and Regeneration Following Ponderosa Pine ...
Guenther, G. E. 1989. Ecological relationships of bitterbrush communities on the Mount Haggin Wildlife Management Area.
Bozeman: Montana State University. 73 p. Thesis.
Guenther, G. E.; Wambolt, C. L.; Frisina, M. R. 1993. Characteristics of bitterbrush habitats that influence canopy cover and
mule deer browsing. Journal of Environmental Management. 36:
175–181.
Harper, K. T.; Buchanan, H. 1983. The ecology of shrubs in Bryce
Canyon National Park with special reference to Purshia tridentata.
In: Tiedemann, A. R.; Johnson, K. L., comps. Proceedings: research and management of bitterbrush and cliffrose in Western
North America. Gen. Tech. Rep. INT-152. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station: 91–98.
Matlock-Cooley, S. J. 1993. Interaction between deermice, cattle
and antelope bitterbrush. Bozeman: Montana State University.
84 p. Thesis.
Meyer, S. E.; Allen, P. S.; Wilson, G. R.; Davis; T. D.; Stevens, R.;
Jorgensen, K.. 1986. Proposed rule for antelope bitterbrush
Purshia tridentata. Association of Official Seed Analysts Newsletter. 69: 21–22.
Meyer, S. E.; Monsen, S. B. 1989. Seed germination biology of
antelope bitterbrush (Purshia tridentata). In: Wallace, A.;
McArthur, E. D.; Haferkamp, M. R., comps. Proceedings: symposium on shrub ecophysiology and biotechnology. Gen. Tech. Rep.
INT-256. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station: 147–157.
Nord, C. C. 1965. Autecology of bitterbrush in California. Ecological
Monographs. 35: 307–334.
Norusis, M. J. 1997. SPSS 7.5 guide to data analysis. New Jersey:
Prentice-Hall. 553 p.
Ott, R. L. 1993. An introduction to statistical methods and data
analysis. Belmont, CA: Duxbury Press. 1051 p.
USDA Forest Service Proceedings RMRS-P-31. 2004
Bedunah, Pflug, and Jones
Peek, J. M.; Johnson, F. D.; Pence, N. N. 1978. Successional trends
in a ponderosa pine/bitterbrush community related to grazing by
livestock, wildlife, and to fire. Journal of Range Management. 31:
49–53.
Pfister, R. D.; Kovalchik, B. L.; Arno, S. F.; Presby, F. C. 1977. Forest
habitat types of Montana. Gen. Tech. Rep. INT-34. Ogden, UT:
U.S. Department of Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station. 174 p.
Shaw, N. L.; Monsen, S. B. 1983. Phenology and growth habits of
nine antelope bitterbrush, desert bitterbrush, Stansbury cliffrose,
and Apache-plume accessions. In: Tiedemann, A. R.; Johnson,
K. L., comps. Proceedings: research and management of bitterbrush and cliffrose in Western North America. Gen. Tech. Rep.
INT-152. Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range Experiment Station:
55–69.
Sherman, R. J.; Chilcote, W. W. 1972. Spatial and chronological
patterns of Purshia tridentata as influenced by Pinus ponderosa.
Ecology. 53: 294–298.
SPSS Inc. 1997. SPSS Base 8.0 syntax reference guide. Software
available from: SPSS Inc., 444 N. Michigan Ave., Chicago, IL
60611.
Tueller, P. T.; Tower, J. D. 1979. Vegetation stagnation in threephase big game exclosures. Journal of Range Management. 32:
258–263.
Young, J. A.; Clements, C. D.; Blank, R. R. 1997. Influence of
nitrogen on antelope bitterbrush seedling establishment. Journal of Range Management. 50: 536–540.
Young, J. A.; Evans, R. A. 1978. Germination requirements as
determinants of species composition of Artemisia rangeland communities. Proceedings, First International Rangelands Congress.
Denver, CO: Society for Range Management: 366–369.
Young, J. A.; Young, C .G. 1986. Collecting, processing and germinating seeds of wildland plants. Portland, OR: Timber Press.
155
Effects of Elevated CO2 on
Growth Responses of Honey
Mesquite Seedlings From Sites
Along a Precipitation Gradient
Justin D. Derner
Charles R. Tischler
H. Wayne Polley
Hyrum B. Johnson
Abstract: We collected seeds from two honey mesquite (Prosopis
glandulosa var. glandulosa) trees at each of three sites along an
east-west precipitation gradient in Texas: Marlin (93.2 cm mean
annual precipitation, MAP), Menard (62.0 cm MAP), and Bakersfield (36.6 cm MAP) to test the hypothesis that CO2 enrichment
would differentially affect plant responses along this gradient.
However, significant interactions between CO2 and the site of origin
were not observed at either harvest (8- or 16-days postemergence).
Growth responses of genotypes of honey mesquite collected from the
east-west precipitation gradient were inconsistent with respect to
the gradient. Significant growth responses to elevated CO2 often
were of small absolute and relative magnitude, especially at the 16day harvest. If genetic differences exist among the genotypes used
in this investigation, they do not affect growth responses to elevated
CO2 under well-watered conditions. Similar enhancement of seedling plant growth to elevated CO2 in genotypes of honey mesquite
from a wide precipitation gradient suggests that this invasive,
woody plant will respond comparably to CO2 enrichment regardless
of precipitation at the site of origin. Whether this genetic potential
is displayed in the field, however, is uncertain as existing environmental conditions may constrain responses to CO2 enrichment. This
invasive species likely will continue to be a problem species on
rangelands in the future.
Introduction ____________________
Honey mesquite (Prosopis glandulosa var. glandulosa) is
an invasive, leguminous shrub that has invaded arid and
semiarid rangeland ecosystems and has become dominant
on much of the southern mixed-grass and shortgrass prairies in the Great Plains (Ansley and others 2001; Archer
1989, 1990, 1995; Van Auken 2000). It has been proposed
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. -Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Justin D. Derner is a Rangeland Scientist, USDA-ARS, High Plains
Grasslands Research Station, 8408 Hildreth Road, Cheyenne, WY 82009,
U.S.A.; FAX: (307) 637-6124; e-mail: jderner@npa.ars.usda.gov. Charles R.
Tischler, H. Wayne Polley, and Hyrum B. Johnson are Plant Physiologist,
Research Leader, and Ecologist, respectively, USDA-ARS, Grassland, Soil &
Water Laboratory, 808 East Blackland Road, Temple, TX 76502, U.S.A.
156
that elevated CO2 may have a significant positive effect on
woody plant species and that this is manifest in invasion by
woody plants and subsequent thickening in grass-dominated ecosystems (Bond and Midgley 2000). Responses of
honey mesquite to elevated CO2 have been primarily conducted in response to soil water conditions (Polley and
others 1996, 1999). Because there is genetic variation within
species in responses to elevated CO2 (Andalo and others
2001; Bezemer and others 1998; Curtis and others 1996;
Klus and others 2001; Schmid and others 1996; Thomas and
Jasienski 1996; Van der Kooij and others 2000), impacts of
CO2 enrichment on honey mesquite plant growth may be
genotype-specific (for example, Mulholland and others 1998;
Roumet and others 1999). This may be particularly important in the context of increasing shrub encroachment of
rangelands, which encompass wide environmental gradients (Polley 1997; Polley and others 1997).
Species like honey mesquite that grow across wide environmental gradients likely differ genetically in ways that
may affect growth responses to elevated CO2. However,
this area of research has received little attention. Three
potential responses to elevated CO2 include (1) similar
responses to CO2 by all genotypes from the environmental
gradient, (2) greater responses to CO2 by genotypes from
more water-limiting environments because of increased
soil water availability associated with elevated CO 2, or
(3) greater responses to CO2 by genotypes from sites with
higher precipitation as they are more rapidly growing
plants.
An underlying assumption of this study is that honey
mesquite differs genetically at different points along an
east-west precipitation gradient in western and central
Texas. We tested the hypotheses that (1) growth of honey
mesquite seedlings grown at ambient CO2 increases as
precipitation at the site of origin increases, and (2) CO2
enrichment preferentially increases growth in genotypes
that already grow most rapidly. Previous investigations
have determined that CO2 enrichment preferentially increased relative growth rate (Poorter 1993) and biomass
(Bunce 1997) in species that had high growth rates. We
addressed these potential responses in this investigation
by growing seedlings of honey mesquite collected from the
precipitation gradient in environmentally controlled glasshouses under optimal temperature and soil water conditions.
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Elevated CO2 on Growth Responses of Honey Mesquite Seedlings From ...
Materials and Methods ___________
The CO2 concentration of air in each of four glasshouse
bays (31∞ 05´N, 97∞ 20´W) was measured at 4-minute
intervals with a Li-Cor Model LI-6262 infrared gas analyzer (Li-Cor, Inc., Lincoln, NE). The CO2 readings were
corrected for atmospheric pressure measured with a Druck
model DPI 260 pressure indicator (Druck, Inc., New
Fairfield, CT). The infrared analyzer was calibrated daily
against four CO2 gas standards and weekly against a LiCor LI-610 dewpoint generator. Air temperature, manually set at 30 ∞C for both day and night in each bay, was
measured in the center of each bay with fine-wire (25 mm
diameter) thermocouples. Pure CO2 gas was injected into
appropriate bays as required to maintain the elevated CO2
concentration. The CO2 concentration of air in the ambient
and elevated CO2 treatments (two bays per treatment)
averaged 366 and 702 mmol mol-1, respectively. Photosynthetic photon flux density (PPFD) was measured on the
glasshouse roof with a point quantum sensor (LI-190SB,
Li-Cor) and within the bays with 1 m long line quantum
sensors (LI-191SA, Li-Cor) mounted about plant height.
On average, the daily integral of PPFD inside the bays was
70 percent of that measured above the glasshouse.
Seeds were collected in August 2000 from two individual
trees (genotypes) of honey mesquite at each of three sites
along a precipitation gradient in Texas: Marlin (93.2 cm
mean annual precipitation, MAP), Menard (62.0 cm MAP),
and Bakersfield (36.6 cm MAP), resulting in a total of six
genotypes. Caution was taken to ensure that these trees
were at least 2 km from any major road, and each was
visually inspected to determine if the morphology of the
trees was typical for the given area.
To prevent confounding influences of seed mass on initial
seedling development (Villar and others 1998), seeds of each
genotype were weighed prior to planting to ensure similar
seed mass within a genotype. Two seeds of each genotype
were planted to depth of 2 cm in each of 32, 0.05 m diameter
by 1.60 m deep pots on July 20, 2001, for a total of 192 pots.
Pots were constructed from polyvinyl chloride pipe cut
longitudinally into two pieces of equal size to facilitate
recovery of intact root systems. The two halves of each pot
were taped together and secured at the base with a perforated cap. Pots were filled with a fine sandy loam soil
(Pedernales series; fine, mixed, thermic typic Paleustalf;
Huckabee and others 1977) with the following properties:
pH = 7.1, organic carbon content = 0.57 percent, 76.2 percent
sand, 16.2 percent silt, 7.6 percent clay, field capacity = 18
percent on a volumetric basis. Soil in the pots was wetted
past field capacity prior to planting by adding one-half
strength Hoagland’s nutrient solution (Hoagland and Arnon
1950). Following planting, pots with each genotype were
randomly assigned to one of four glasshouse bays, two
maintained at ambient CO2 and two at elevated CO2, with
eight pots per genotype in each bay. Full strength Hoagland’s
solution was added daily to each pot to maintain soil water
content near field capacity. Seedlings were randomly thinned
to one per pot following emergence.
One-half of the total number of seedlings were harvested
at each harvest (8- or 16-days postemergence). Aboveground
biomass was separated into stem and leaf components with
USDA Forest Service Proceedings RMRS-P-31. 2004
Derner, Tischler, Polley, and Johnson
the area of each leaf blade measured using a LI-3000A
portable leaf area meter (Li-Cor, Inc., Lincoln, NE). Soil was
manually washed from roots and the depth of deepest root
penetration recorded. Roots were digitally scanned at high
resolution (600 dpi) using the WinRHIZO software (Regent
Instruments, Inc., Quebec, Canada, version 4.1c) and hardware (Hewlett Packard ScanJet 6100C scanner) to determine root length, surface area, and root volume. Roots were
not stained prior to analyses, which resulted in underestimations (Bouma and others 2000), but these were minimized by using the WinRHIZO automatic threshold (that is,
Lagarde’s method) for pale roots with greater sensitivity.
Aboveground tissues and roots were dried at 60 ºC for 72
hours prior to weighing.
Data were analyzed using ANOVA by harvest date using
a split-split plot design with CO2 as the split plot and the site
of seed origin as the split-split plot in a balanced design (SAS
Institute, Inc. 1994). Single degree freedom contrasts were
used to compare the two genotypes within each site of seed
origin. Means were separated by Duncan’s multiple range
test at the 0.10 level of significance. When needed to normalize residuals, data were logarithmically transformed before
analysis; means and standard errors are reported after back
transforming.
Results ________________________
Aboveground, but not belowground, variables of honey
mesquite seedlings at 8 days postemergence were influenced
by both the site of genotype origin and elevated CO2 (table 1).
However, significant interactions between the site of origin
and CO2 were not observed, indicating that growth responses to elevated CO2 were similar for mesquite seedlings
from along the precipitation gradient. Aboveground variables, with the exception of stem mass, were 20 to 110
percent greater for the genotypes from the site with greatest
MAP compared to the two lower MAP sites, both of which did
not significantly differ for the majority of aboveground
responses. Belowground variables did not differ with respect
to the site of genotype origin. There were no significant
differences for aboveground or belowground growth responses
between the two genotypes from each site of collection (data
not shown). Elevated CO2 increased all aboveground variables, but responses were significant for only other leaf area
(excluding cotyledon and leaf pair 1) and mass, total leaf
mass, and aboveground mass. Total plant mass was similar
between CO2 levels, however, as decreased root mass with
elevated CO2 offset the increased aboveground mass. In
contrast to aboveground variables, belowground responses
were generally reduced with elevated CO2, but no significant
effects were detected.
In contrast to observed results from the 8-day harvest,
both changes in aboveground and belowground variables of
honey mesquite seedlings at 16 days postemergence were
influenced by the site of genotype origin and elevated CO2
(table 2). Similar to the 8-day harvest, no significant interactions between CO2 and the site of genotype origin were
observed. Above- and belowground growth responses to the
site of genotype origin decreased in magnitude at the 16-day
harvest compared to the 8-day harvest. In general, growth
responses were greater for genotypes from the wettest than
157
Derner, Tischler, Polley, and Johnson
Effects of Elevated CO2 on Growth Responses of Honey Mesquite Seedlings From ...
Table 1—Mean (± 1 SE) aboveground and belowground responses at 8-days postemergence for honey mesquite seedlings of genotypes from along
an east-west precipitation gradient in Texas (Marlin [93.2 cm mean annual precipitation, MAP], Menard [62.0 cm MAP] and Bakersfield
[36.6 cm MAP]) exposed to ambient and elevated CO2 concentrations (366 and 702 mmol mol-1, respectively).
CO2
Variable
Ambient
Elevated
93.2
MAP at site of genotype origina
62.0
36.6
- - - - - - - - - - - - - - - - - - - - - - cm - - - - - - - - - - - - - - - - - - - - Aboveground
Cotyledon leaf area (cm2)
Cotyledon mass (g)
Leaf pair 1 leaf area (cm2)
Leaf pair 1 mass (g)
Other leaf area (cm2)
Other leaf mass (g)
Stem mass (g)
Total leaf area (cm2)
Total leaf mass (g)
Aboveground mass (g)
2.13 (0.13)
0.0144 (0.0009)
2.27 (0.15)
0.0080 (0.0015)
2.28 (0.21)
0.0093 (0.0009)
0.0089 (0.0008)
6.68 (0.40)
0.0317 (0.0018)
0.0406 (0.0022)
2.27 (0.26)
0.0158 (0.0021)
2.34 (0.34)
0.0091 (0.0014)
2.86 (0.22)b
0.0176 (0.0048)b
0.0111 (0.0018)
7.48 (1.03)
0.0425 (0.0065)b
0.0537 (0.0077)b
2.85 (0.26)a
0.0180 (0.0025)a
2.82 (0.28)a
0.0101 (0.0015)a
3.31 (0.63)a
0.0204 (0.0059)a
0.0115 (0.0020)a
8.97 (1.07)a
0.0485 (0.0075)a
0.0599 (0.0089)a
1.89 (0.20)b
0.0149 (0.0014)b
1.76 (0.33)b
0.0069 (0.0008)c
1.91 (0.27)b
0.0081 (0.0013)b
0.0094 (0.0022)a
5.56 (0.73)b
0.0299 (0.0032)b
0.0393 (0.0048)b
1.85 (0.10)b
0.0128 (0.0009)b
2.19 (0.22)b
0.0081 (0.0009)b
2.27 (0.24)b
0.0097 (0.0011)b
0.0089 (0.0010)a
6.31 (0.45)b
0.0305 (0.0019)b
0.0394 (0.0023)b
Belowground
Root depth (cm)
Root length (cm)
Root surface area (cm2)
Root volume (m3)
Root mass (g)
51.6 (6.6)
90.1 (12.2)
17.5 (1.8)
0.26 (0.03)
0.0151 (0.0013)
49.6 (4.8)
77.7 (7.8)
14.8 (1.3)
0.21 (0.02)
0.0135 (0.0016)
49.8 (6.4)a
92.2 (8.6)a
17.4 (1.4)a
0.25 (0.02)a
0.0149 (0.0018)a
46.4 (10.8)a
75.2 (14.4)a
15.3 (2.7)a
0.23 (0.04)a
0.0132 (0.0025)a
53.5 (6.3)a
84.0 (14.4)a
16.1 (2.2)a
0.23 (0.03)a
0.0147 (0.0014)a
Whole plant
Total plant mass (g)
0.0557 (0.0027)
0.0672 (0.0086)
0.0749 (0.0097)a
0.0524 (0.0067)b
0.0541 (0.0028)b
a
b
Different letters among the MAP levels indicate a significant difference between means.
Significant (P < 0.10) difference between ambient and elevated CO2 means.
Table 2—Mean (± 1 SE) aboveground and belowground responses at 16-days postemergence for honey mesquite seedlings of genotypes from along
an east-west precipitation gradient in Texas (Marlin [93.2 cm mean annual precipitation, MAP], Menard [62.0 cm MAP] and Bakersfield
[36.6 cm MAP]) exposed to ambient and elevated CO2 concentrations (366 and 702 mmol mol-1, respectively).
CO2
Variable
Ambient
Elevated
93.2
MAP at site of genotype origina
62.0
36.6
- - - - - - - - - - - - - - - - - - - - - - cm - - - - - - - - - - - - - - - - - - - - Aboveground
Cotyledon leaf area (cm2)
Cotyledon mass (g)
Leaf pair 2&3 leaf area (cm2)
Leaf pair 2&3 mass (g)
Other leaf area (cm2)
Other leaf mass (g)
Stem mass (g)
Total leaf area (cm2)
Total leaf mass (g)
Aboveground mass (g)
2.39 (0.20)
1.7553 (0.0011)
6.02 (0.40)
1.7706 (0.0022)
9.64 (0.97)
1.7826 (0.0047)
1.7652 (0.0040)
18.06 (1.48)
5.3085 (0.0074)
7.0737 (0.0097)
2.57 (0.15)
1.7585 (0.0010)b
7.18 (0.39)b
1.7793 (0.0026)b
10.18 (1.20)
1.7901 (0.0062)
1.7756 (0.0036)
19.93 (1.54)
5.3280 (0.0086)b
7.1036 (0.0116)b
2.93 (0.23)a
1.7594 (0.0012)a
7.38 (0.45)a
1.7778 (0.0023)a
12.65 (1.52)a
1.7985 (0.0076)a
1.7743 (0.0064)a
22.96 (2.07)a
5.3358 (0.0103)a
7.1100 (0.0146)a
2.29 (0.20)b
1.7569 (0.0015)b
6.71 (0.55)a
1.7779 (0.0044)a
8.70 (0.81)b
1.7849 (0.0051)a
1.7674 (0.0035)a
17.70 (1.37)b
5.3196 (0.0101)a
7.0870 (0.0130)a
2.08 (0.13)b
1.7535 (0.0011)b
5.44 (0.40)b
1.7680 (0.0023)b
7.60 (0.57)b
1.7723 (0.0025)b
1.7666 (0.0027)a
15.13 (0.72)b
5.2939 (0.0044)b
7.0605 (0.0067)b
Belowground
Root depth (cm)
Root length (cm)
Root surface area (cm2)
Root volume (m3)
Root mass (g)
96.2 (5.7)
201.2 (18.4)
32.5 (2.6)
0.44 (0.04)
1.7861 (0.0044)
101.3 (5.0)
240.3 (20.4)
40.1 (2.5)b
0.57 (0.04)b
1.8024 (0.0055)b
107.2 (6.8)a
250.9 (25.2)a
40.1 (3.4)a
0.55 (0.05)a
1.7998 (0.0063)a
93.9 (7.2)b
171.5 (17.4)b
31.5 (3.1)b
0.46 (0.06)a
1.7955 (0.0085)a
91.9 (5.3)b
218.0 (21.4)ab
34.5 (2.8)ab
0.45 (0.04)a
1.7844 (0.0038)a
Whole plant
Total plant mass (g)
8.8598 (0.0135)
8.9059 (0.0164)b
8.9098 (0.0135)a
8.8825 (0.0210)a
8.8449 (0.0102)b
a
b
Different letters among the MAP levels indicate a significant difference between means.
Significant (P < 0.10) difference between ambient and elevated CO2 means.
158
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Elevated CO2 on Growth Responses of Honey Mesquite Seedlings From ...
driest site. Consistent with the 8-day harvest, there were no
significant differences for aboveground or belowground
growth responses between the two genotypes for each site of
collection (data not shown). Elevated CO2 increased all
aboveground responses, but significant differences were
observed for only leaf area and mass of leaf pairs 2 and 3,
total leaf mass, and aboveground mass. The magnitude of
these CO2-induced growth responses was much reduced
compared to the 8-day harvest. Belowground responses
were greater with elevated CO2, with significant increases
at elevated CO2 in root surface area, volume, and mass.
Discussion _____________________
Growth responses of honey mesquite genotypes collected
from the east-west precipitation gradient were inconsistent
with respect to the gradient, suggesting that our underlying
assumption that honey mesquite differs genetically at different points along an east-west precipitation gradient in
western and central Texas was incorrect, at least under wellwatered conditions. In addition, this finding does not support our hypothesis that growth of seedlings grown at
ambient CO2 increases as precipitation at the site of origin
increases. In general, genotypes from the wettest site (Marlin) had more leaf area, more mass, and greater root growth
than genotypes from the driest site (Bakersfield), but growth
responses of genotypes from the site with intermediate
precipitation (Menard), however, were primarily responsible for the inconsistent growth response along the gradient. Growth responses of seedlings from this site were
similar to those of genotypes from the wettest site for some
variables and similar to genotypes from the driest site for
other variables at the 16-day harvest. This inconsistency
merits additional research attention, with a need to investigate additional genotypes from each site.
Elevated CO2 increased aboveground variable responses
at both harvest dates, while belowground variable responses
to elevated CO2 were inconsistent with decreases occurring
at the 8-day harvest and increases at the 16-day harvest.
Most of these responses, however, were not significant.
Significant differences in response to elevated CO2 often
were of small absolute and relative magnitude, especially at
the 16-day harvest. Significant interactions between CO2
and the site of genotype origin were not observed for either
the 8-day or 16-day harvest, indicating that observed growth
responses to elevated CO2 were similar for mesquite seedlings regardless of MAP at site of origin. Therefore, our
hypothesis that elevated CO2 would preferentially increase
growth of the more rapidly growing plants from the sites
with high than low precipitation was not supported for
above- or belowground variables at either harvest date. If
genetic differences exist among the genotypes used in this
investigation, they do not affect growth responses to elevated
CO2 under well-watered conditions.
Inconsistent growth responses of aboveground and
belowground variables to elevated CO2 at the 8- and 16-day
harvests are likely attributable to relative sink strengths and
carbon allocation patterns of these honey mesquite seedlings.
Increased aboveground, but decreased belowground, responses to elevated CO2 at the 8-day harvest conform to
priority of establishing sufficient photosynthetic machinery
USDA Forest Service Proceedings RMRS-P-31. 2004
Derner, Tischler, Polley, and Johnson
for seedling survival at the expense of root growth. The
reduction in magnitude of aboveground response to elevated
CO2 at the 16-day harvest, and the associated increase in
belowground response, indicate that photosynthetic machinery was quickly established and carbon allocation priorities shifted more belowground following the first harvest.
Elevated CO2 did not affect partitioning of dry matter between shoot and root of the grass Dactylis glomerata at high
nitrogen, but at low nitrogen greater partitioning was observed into the shoot during early stages of growth (Harmens
and others 2000). This suggests that in spite of daily additions of full-strength Hoaglands’s solution, the inherently
low nitrogen status and high leaching potential of the sandy
loam soil resulted in low nitrogen availability to honey
mesquite seedlings. Under limiting nutrient conditions,
plant growth under elevated CO2 is often negligible
(Bernacchi and others 2000; Kimball and others 2002; but
see Lloyd and Farquhar 1996).
Similar enhancement of early seedling plant growth in
genotypes of honey mesquite from a wide precipitation
gradient to elevated CO2 suggests that this invasive, woody
plant will respond comparably to CO2 enrichment regardless of precipitation at the site of origin. Whether this genetic
potential is displayed in the field, however, is uncertain, as
existing environmental conditions may constrain responses
to CO2 enrichment. This invasive species likely will continue
to be a problem species on rangelands in the future and may
even increase competitiveness at the expense of warmseason grasses, which currently dominate the southern
mixed-grass and shortgrass prairies, as these grasses are
generally not as responsive to elevated CO2 as is honey
mesquite (Polley and others 1994).
Acknowledgments ______________
Alicia Naranjo, Kyle Tiner, Holly Harland, Brooke Kramer,
Jenny Fikes, and Dustin Coufal assisted with data collection
and entry. Chris Kolodziejczyk maintained CO2 and environmental control systems and data records.
References _____________________
Andalo, C.; Goldringer, I.; Godelle, B. 2001. Inter- and intragenotypic
competition under elevated carbon dioxide in Arabidopsis thaliana.
Ecology. 82: 157–164.
Ansley, J. R.; Wu, X. B.; Kramp, B. A. 2001. Observation: long-term
increases in mesquite canopy cover in a north Texas savanna.
Journal of Range Management. 54: 171–176.
Archer, S. 1989. Have southern Texas savannas been converted to
woodlands in recent history? The American Naturalist. 134:
545–561.
Archer, S. 1990. Development and stability of grass/woody mosaics
in a subtropical savanna parkland, Texas, U.S.A. Journal of
Biogeography. 17: 453–462.
Archer, S. 1995. Tree-grass dynamics in a Prosopis-thornscrub
savanna parkland: reconstructing the past and predicting the
future. Ecoscience. 2: 83–99.
Bernacchi, C. J.; Coleman, J. S.; Bazzaz, F. A.; McConnaughay, K.
D. M. 2000. Biomass allocation in old-field annual species grown
in elevated CO2 environments: no evidence for optimal partitioning. Global Change Biology. 6: 855–863.
Bezemer, T. M.; Thompson, L. J.; Jones, T. H. 1998. Poa annua
shows inter-generational differences in response to elevated CO2.
Global Change Biology. 4: 687–691.
159
Derner, Tischler, Polley, and Johnson
Bond, W. J.; Midgley, G. F. 2000. A proposed CO2-controlled mechanism of woody plant invasion in grasslands and savannas. Global
Change Biology. 6: 865–869.
Bouma, T. J.; Nielsen, K. L.; Koutstall, B. 2000. Sample preparation
and scanning protocol for computerized analysis of root length
and diameter. Plant and Soil. 218: 185–196.
Bunce, J. A. 1997. Variation in growth stimulation by elevated
carbon dioxide in seedlings of some C3 crop and weeds species.
Global Change Biology. 3: 61–66.
Curtis, P. S.; Klus, D. J.; Kalisz, S.; Tonsor, S. J. 1996. Intraspecific
variation in CO2 responses in Raphanus raphanistrum and
Plantago lancelata: assessing the potential for evolutionary change
with rising atmospheric CO2. In: Korner, C.; Bazzaz, F. A., eds.
Carbon dioxide, populations, and communities. London: Academic Press: 13–22.
Harmens, H.; Stirling, C. M.; Marshall, C.; Farrar, J. F. 2000. Is
partitioning of dry weight and leaf area within Dactylis glomerata
affected by N and CO2 enrichment? Annals of Botany. 86: 833–839.
Hoagland, D. R.; Arnon, D. I. 1950. The water-culture method for
growing plants without soil. California Agricultural Experiment
Station Circular. 347: 1–39.
Huckabee, J. W., Jr.; Thompson, D. R.; Wyrick, J. C.; Paulat, E. G.
1977. Soil survey Bell County Texas. Washington, DC: U.S.
Department of Agriculture, Soil Conservation Service.
Kimball, B. A.; Kobayashi, K.; Bindi, M. 2002. Responses of agricultural crops to Free-Air CO2 Enrichment. Advances in Agronomy.
77: 293–368.
Klus, D. J.; Kalisz, S.; Curtis, P. S.; Terri, J. A.; Tonsor, S. J. 2001.
Family- and population-level responses to atmospheric CO2 concentration: gas exchange and the allocation of C, N, and biomass
in Plantago lanceolata (Plantaginaceae). American Journal of
Botany. 88: 1080–1087.
Lloyd, J.; Farquhar, G. D. 1996. The CO2 dependence of photosynthesis, plant growth responses to elevated CO2 concentrations
and their interactions with soil nutrient status. I. General principles and forest ecosystems. Functional Ecology. 10: 4–32.
Mulholland, B. J.; Craigon, J.; Black, C. R.; Colls, J. J.; Atherton, J.;
Landon, G. 1998. Growth, light interception and yield responses
of spring wheat (Triticum aestivum L.) grown under elevated CO2
and O3 in open-top chambers. Global Change Biology. 4: 121–130.
Polley, H. W. 1997. Invited synthesis paper: implications of rising
atmospheric carbon dioxide concentration for rangelands. Journal of Range Management. 50: 562–577.
160
Effects of Elevated CO2 on Growth Responses of Honey Mesquite Seedlings From ...
Polley, H. W.; Johnson, H. B.; Mayeux, H. S. 1994. Increasing CO2:
comparative responses of the C4 grass Schizachyrium and grassland invader Prosopis. Ecology. 75: 976–988.
Polley, H. W.; Johnson, H. B.; Mayeux, H. S.; Tischler, C. R.; Brown,
D. A. 1996. Carbon dioxide enrichment improves growth, water
relations and survival of droughted honey mesquite (Prosopis
glandulosa) seedlings. Tree Physiology. 16: 817–823.
Polley, H. W.; Mayeux, H. S.; Johnson, H. B.; Tischler, C. R. 1997.
Viewpoint: atmospheric CO2 soil water, and shrub/grass ratios on
rangelands. Journal of Range Management. 50: 278–284.
Polley, H. W.; Tischler, C. R.; Johnson, H. B.; Pennington, R. E.
1999. Growth, water relations, and survival of drought-exposed
seedlings from six maternal families of honey mesquite (Prosopis
glandulosa): responses to CO2 enrichment. Tree Physiology. 19:
359–366
Poorter, H. 1993. Interspecific variation in the growth response of
plants to an elevated ambient CO2 concentration. Vegetatio. 104/
105: 77–97.
Roumet, C.; Laurnet, G.; Roy, J. 1999. Leaf structure and chemical
composition as affected by elevated CO2: genotypic responses of
two perennial grasses. New Phytologist. 143: 73–81.
SAS Institute, Inc. 1994. SAS/STAT user’s guide, release 6.03, 4th
ed. Cary, NC: SAS Institute, Inc.
Schmid, B.; Birrer, A.; Lavigne, C. 1996. Genetic variation in the
response of plant populations to elevated CO2 in a nutrient-poor,
calcareous grassland. In: Korner, C.; Bazzaz, F. A., eds. Carbon
dioxide, populations, and communities. London: Academic Press:
31–50.
Thomas, S. C.; Jasienski, M. 1996. Genetic variability and the
nature of microevolutionary responses to elevated CO2. In:
Korner, C.; Bazzaz, F. A., eds. Carbon dioxide, populations, and
communities. London: Academic Press: 51–81.
Van Auken, O. W. 2000. Shrub invasions of North American semiarid grasslands. Annual Review of Ecological Systems. 31: 197–215.
Van der Kooij, T. A. W.; De Kok, L. J.; Stulen, I. 2000. Intraspecific
variation in the response of Arabidopsis thaliana lines to elevated
atmospheric CO2. Phyton-annales Rei Botanicae. 40: 125–132.
Villar, R.; Veneklaas, E. J.; Jordano, P.; Lambers, H. 1998. Relative
growth rate and biomass allocation in 20 Aegilops (Poaceae)
species. New Phytologist. 140: 425–437.
USDA Forest Service Proceedings RMRS-P-31. 2004
Responses of Seedlings of Five
Woody Species to Carbon
Dioxide Enrichment
Charles R. Tischler
Justin D. Derner
H. Wayne Polley
Hyrum B. Johnson
Abstract: Encroachment of woody species into formerly productive
rangeland is an immense problem in the Southwest. Effects of
predicted global change scenarios on seedlings of such species are
largely unstudied. Seedlings of five invasive woody legume species
(honey mesquite [Prosopis glandulosa], huisache [Acacia farnesiana],
honey locust [Gleditsia triacanthos], Eve’s necklace [Sophora affinis],
and Paloverde [Parkinsonia aculeata]) were grown for 23 days in
glasshouses in Temple, TX, at ambient and twice ambient levels of
atmospheric carbon dioxide. Plants were kept well watered and
fertilized. Seedlings of all species responded positively to CO2
enrichment, with significant differences in seedling mass observed
1 week after emergence. Mesquite and huisache responded most
strongly to elevated CO2. Other work from our laboratory suggests
that at least for honey mesquite and huisache, seedling mortality is
near zero beyond 3 weeks postemergence, even under water stress
conditions. Thus, these invasive species will likely become even
more problematic in the future as atmospheric CO2 increases.
Introduction ____________________
It is now widely accepted that atmospheric CO2 concentration has increased in the recent past and will continue
to increase (Trabalka and others 1986). An environment
enriched in CO2 has direct stimulatory effects on plants
with C3 photosynthetic carbon metabolism (Bunce 1997),
although moderate effects, as well as indirect effects, are
noted in C4 plants (Owensby and others 1993). Because
most of the grasslands in the Southwestern United States
consist of C4 species that have been or are currently being
invaded by C3 woody species (Polley and others 1996),
effects of this atmospheric change on growth and development of brush species merits study.
For most invasive species, success in the seedling stage is
a critical factor influencing eventual success of that species.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Charles R. Tischler, H. Wayne Polley, and Hyrum B. Johnson are Plant
Physiologist, Research Leader and Ecologist, respectively, USDA-ARS Grassland, Soil & Water Research Laboratory, 808 E. Blackland Road, Temple, TX
76502, U.S.A.; FAX: (254) 770-6561; e-mail: ctischler@spa.ars.usda.gov. Justin D. Derner is a Range Scientist, USDA-ARS, High Plains Grasslands
Research Station, 8408 Hildreth Road, Cheyenne, WY 82009, U.S.A.
USDA Forest Service Proceedings RMRS-P-31. 2004
One reason for this is that woody species tend to be long-lived
perennials that are avoided by livestock, and relatively
resistant to environmental extremes, insects and pathogens, and other common causes of mortality. It has been
proposed that elevated CO2 may have a significant positive
effect on woody plant species and that this is manifest in
invasion by woody plants and subsequent thickening in
grass-dominated ecosystems (Bond and Midgley 2000).
We studied effects of a twice ambient CO2 atmosphere on
the growth and development of five invasive woody legumes
for the first 23 days after emergence in these experiments.
These species differ markedly in invasive potential, and we
hypothesize that positive effects of CO2 would be most
pronounced on the most invasive species. Mesquite (Prosopis
glandulosa) and huisache (Acacia farnesiana) are found in
almost pure stands on millions of acres of rangeland in
Texas, with mesquite being cosmopolitan but more prevalent in the western portion of south-central and southwestern Texas, and huisache being more prevalent in southcentral Texas and the eastern portion of the Rio Grande
Plains, but not extending north beyond central Texas.
Paloverde (Parkinsonia aculeata), honey locust (Gleditsia
triacanthos), and Eve’s necklace (Sophora affinis) are rare to
common across the eastern half of Texas, generally never
existing in pure stands, but being minor components of
mixed woodlands. Occasionally Eve’s necklace and Paloverde
are used as ornamentals.
Materials and Methods ___________
Seed of the five species used in this experimentation
were harvested and scarified during the summer of 1996.
Individual seed were weighed so that a narrow range of
seed masses could be selected for each species, thus insuring relatively consistent cotyledonary leaf area for each
species. The range of seed masses used for each species
were Eve’s Necklace 0.010 to 0.0800 g; Honey locust 0.0801
to 0.0900 g; Paloverde 0.1110 to 0.1160 g; mesquite 0.0451
to 0.0500 g; and huisache: 0.0651 to 0.0700 g. Planting
dates were staggered to synchronize seedling emergence
among species. Eve’s necklace seed were planted on August
19, Honey Locust on August 26, and Paloverde, Mesquite,
and Huisache were planted on August 27, with emergence
for all species occurring on August 30 and 31, 1996.
161
Tischler, Derner, Polley, and Johnson
Responses of Seedlings of Five Woody Species to Carbon Dioxide Enrichment
Seed were planted at a depth of 1.5 cm in commercial
potting mix in tapered plastic cones 25 cm deep and 6.25 cm
in diameter. Total cone volume was 425 cm3. Cones were
watered at planting and weekly with full strength Hoagland’s
solution, and with tap water as necessary. As seedlings
emerged, cones were transferred to interlocking racks, each
with a capacity of 20 cones. Species were randomized within
a rack, with four individuals of each species in a rack.
Two adjacent glasshouse bays at the Grassland, Soil, and
Water Research Laboratory (Temple, TX) were utilized;
CO2 concentrations were 365 and 700 ppm. Air within each
bay was sampled each 4 minutes with a Li-Cor Model LI622262 infrared gas analyzer to determine CO2 concentration. Pure CO2 was injected into the 700 ppm glasshouse
bay as needed to maintain the desired concentration, while
in the 365 ppm bay, outside air was circulated through the
glasshouse to maintain ambient conditions. Air temperature was maintained near that of outdoor temperature. The
late summer of 1996 was dry with few clouds, and thus the
seedlings received saturating levels of light most of the
time.
Destructive harvests were made on days 3, 8, 13, 18, and
23. Root and shoot dry mass, and cotyledonary and true leaf
dry mass and area were determined for seven plants of each
species at each harvest. Data was analyzed by standard
ANOVA techniques, with comparisons made only within a
species and harvest date.
cm2 at day 13, compared to 2.6 cm2 for mesquite at day 13,
the next greatest area—both values for ambient atmosphere). Because of different developmental strategies between the woody species and differences in rates of development, detailed comparisons between the species are not
meaningful.
Results ________________________
Cotyledonary Mass
On days 8 and 13, some significant effects of elevated CO2
on cotyledonary mass were observed (table 1). These effects
were not observed later in development, probably because
of the slow progression of the cotyledonary leaves to senescence, and the weight loss experienced by cotyledonary
leaves before they abscise. With the exception of Paloverde,
cotyledonary mass of all other species was at near maximum at day 3. This is because Paloverde is the only species
that has significant expansion of cotyledonary leaves (6.13
Table 1—Cotyledonary masses of five woody species at five sampling
dates.
Entry
Paloverde 350
Paloverde 700
Locust 350
Locust 700
Mesquite 350
Mesquite 700
Huisache 350
Huisache 700
Eve's necklace 350
Eve's necklace 700
Day 3
Cotyledon massa
Day 8
Day 13 Day 18 Day 23
- - - - - - - - - - - - - - - mg - - - - - - - - - - - - - - - 0.028a 0.041b 0.050a 0.044b 0.043a
.026a
.057a
.057a
.053a
.051a
.016a
.012b
.010b
.010a
.010a
.016a
.015a
.011a
.009a
.010a
.018a
.023b
.021b
.017b
.018a
.018a
.031a
.038a
.033a
.027a
.013a
.010b
.008a
.006a
.003a
.014a
.014b
.008a
.005a
.004a
.015a
.012a
.015a
.016a
.015a
.018a
a
Values for a species in a column followed by a different letter are significantly
different (P < 0.05), Duncan’s Multiple Range Test.
162
Cotyledonary Leaf Area and True Leaf
Mass and Area
When statistical differences were noted, cotyledonary leaf
area and true leaf mass and area were greater for species in
elevated versus ambient CO2 atmospheres (data not presented).
Because elevated CO2 modifies plant phenology, we did not
attempt to make comparisons of specific leaf weights,
because these comparisons are not instructive.
Root Mass
By day 23, root mass for all species was numerically higher
at elevated CO2 (table 2). None of these differences reached
statistical significance because of the great deal of variation
in root mass from plant to plant. Larger sample sizes would
be required to establish statistical significance.
Total Biomass
At day 23, total biomass was greater in every case for
plants in enriched CO2, although this difference was significant only for mesquite (table 3). However, we observed
significant differences for Paloverde at days 8, 13, and 18;
for Locust at days 8 and 18; for Mesquite additionally at
days 3, 8, 13, and 18; for Huisache at days 8, 13, and 18, and
for Eve’s Necklace at day 13. These data suggest an overall
significant effect of CO2 on biomass accumulation, but
point out a complicating factor often overlooked by investigators. This is the effect of “rests” or temporary lapses in
growth (often concomitant with increases in complexity) as
described by Stebbins (1976) and Vogel (1980). Comparisons between species become meaningless if one species is
in a “rest” stage while a second is actively growing, as
Stebbins (1976) demonstrated.
Table 2—Root dry mass of five woody species at five sampling dates.
Entry
Day 3
Paloverde 350
Paloverde 700
Locust 350
Locust 700
Mesquite 350
Mesquite 700
Huisache 350
Huisache 700
Eve's necklace 350
Eve's necklace 700
Day 8
Root massa
Day 13 Day 18 Day 23
- - - - - - - - - - - - - - - mg - - - - - - - - - - - - - - - 0.017a 0.036a 0.043a
0.062a 0.081a
.016a
.034a
.050a
.067a .087a
.016a
.016b
.029a
.049a .067a
.015a
.020a
.034a
.054a .075a
.010b
.015a
.017b
.033b .047a
.012a
.016a
.023a
.046a .070a
.012a
.013a
.017a
.029b .040b
.012a
.249a
.023a
.039b .058a
.013a
.017a
.041a
.013a
.023a
.045a
a
Values for a species in a column followed by a different letter are significantly
different (P < 0.05), Duncan’s Multiple Range Test.
USDA Forest Service Proceedings RMRS-P-31. 2004
Responses of Seedlings of Five Woody Species to Carbon Dioxide Enrichment
Tischler, Derner, Polley, and Johnson
Table 3—Total dry mass of five woody species at five sampling dates.
Species
CO2
Paloverde
Paloverde
Locust
Locust
Mesquite
Mesquite
Huisache
Huisache
Eve’s necklace
Eve’s necklace
350
700
350
700
350
700
350
700
350
700
Day 3
Day 8
Total massa
Day 13
Day 18
Day 23
- - - - - - - - - - - - - - - - - - - - - mg - - - - - - - - - - - - - - - - - - - - 0.051b
0.077b
0.134b
0.166b
0.240a
.048a
.111a
.168a
.230a
.267a
.039a
.049b
.079a
.118b
.167a
.039a
.070a
.093a
.139b
.189a
.031b
.049b
.070b
.123b
.167b
.035a
.062a
.096a
.173a
.295a
.028a
.042b
.059b
.091B
.125a
.031a
.053a
.088a
.129a
.188a
.033a
.058b
.153a
.035a
.089a
.199a
a
Values for a species in a column followed by a different letter are significantly different (P < 0.05), Duncan’s
Multiple Range Test.
Conclusions ____________________
Our data for five woody species vividly illustrate positive
growth effects of elevated CO2 early in the development of
species. However, our glasshouse results probably grossly
underestimate the effects of CO2 in field or range situations.
By increasing water use efficiency of all species in an
ecosystem, elevated CO2 alters water balance of the system,
thus favoring seedling establishment. Although no data
exists on this topic, it would also be logical to expect a
positive influence of elevated CO2 on reproductive output of
shrub and brush species, thus enriching the seedbank of
these invaders.
Obviously, cumulative effects of increasing CO2 on rangeland ecosystems is an experiment in progress, whose result
will not be completely chronicled for centuries. However, the
data we present suggests that increasing CO2 will favor shrub
and brush invasion at the expense of grass communities.
References _____________________
Bond, W. J.; Midgley, G. F. 2000. A proposed CO2-controlled mechanism of woody plant invasion in grasslands and savannas. Global
Change Biology. 6: 865–869.
Bunce, J. A. 1997 Variation in growth stimulation by elevated
carbon dioxide in seedlings of some C3 crop and weed species.
Global Change Biology. 3: 61–66.
USDA Forest Service Proceedings RMRS-P-31. 2004
Fenner, M. 1983. Relationships between seed weight, ash content
and seedling growth in twenty-four species of Compositae. New
Phytology. 95: 697–706.
Owensby, C. E.; Coyne, P. I.; Ham, J. M.; Auen, L. M.; Knapp, A. K.
1993. Biomass production in a tallgrass prairie ecosystem exposed to ambient and elevated CO2. Ecological Applications:
3(4): 644–653.
Polley, H. W.; Johnson, H. B.; Mayeux, H. S.; Tischler, C. R. 1996.
Are some of the recent changes in grassland communities a
response to rising CO2 concentrations? In: Korner, Christian;
Bazzaz, Fakhri A., eds. Carbon dioxide, populations, and communities. New York: Academic Press: 177–195.
Stebbins, G. L. 1976. Seed and seedling ecology in annual legumes.
I. A comparison of seed size and seedling development in some
annual species. Oecologia Plantarum. 11: 321–331.
Tischler, C .R.; Polley, H. W.; Johnson, H. B.; Mayeux, H. S. 1998.
Environment and seedling age influence mesquite response to
epicotyl removal. Journal of Range Management. 51: 361–365.
Tischler, C. R.; Polley, H. W.; Johnson, H. B.; Pennington, R. E.
2000. Seedling response to elevated CO2 in five epigeal species.
International Journal of Plant Science. 161: 779–783.
Trabalka, J. R.; Edmonds, J. A.; Reilly, J. M.; Gardner, R. H.;
Reichle, D. E. 1986 Atmospheric CO2 projections with globally
averaged carbon cycle models. In: Trabalka, J. R.; Reichle, D. E.,
eds. The changing carbon cycle: a global analysis. New York:
Springer-Verlag: 534–560.
Vogel, de, E. F. 1980. Seedlings of dicotyledons. Pudoc, Centre for
Agricultural Publishing and Documentation, Wageningen.
163
Soil Moisture Attributes of Three
Inland Sand Dunes in the Mojave
Desert
Simon A. Lei
Abstract: Soil moisture attributes were investigated in three active
sand dunes in the Mojave Desert of southern Nevada and California.
Soil water content increased significantly with increasing soil
depth, and decreased when moving toward dune habitats. Path
analysis revealed direct causal effects of soil depth and habitat type
(dune and nondune) on soil water content. Among soil moisture
variables, significant interaction was detected between habitat type
and geomorphic surface (terrace and slope) for area of water spread
(surface water runoff). When examining habitat type or geomorphic
surface alone, significant differences were detected in all measured
moisture variables. Significant differences were also found in site
for water infiltration and depth of water penetration. The Kelso,
Ash Meadows, and Death Valley dunes had different soil moisture
attributes compared with adjacent Larrea-Ambrosia shrublands.
Introduction ____________________
Active sand dunes cover less than 1 percent of the North
American deserts (Sharp 1966). Desert sand dunes are
edaphically distinct from surrounding habitats in terms of
extremely high percentage of sand, instability of sand substrates, and rate of sand movement. Long and dry seasons,
low soil nutrient levels, high air and soil temperatures, as
well as mobile and abrasive sand with coarse texture are
characteristics of inland desert dunes (Bowers 1986; Pavlik
1985). Low soil moisture, low organic litter, and matter are
also typical of dune environments (Barbour and others 1999;
Bowers 1982). However, soil moisture increases with depth
in the inland dunes (Bowers 1982; Prill 1968; Sharp 1966).
Although sand dune ecosystems appear simple, they are
actually complex and delicately balanced (Bowers 1986).
Sand dunes are unique continental islands isolated by their
physical and biological properties or by factors related to the
evolution of the landscape through geological time (Brown
1972). Because of low economic and agricultural values,
dunes are one of the least studied ecosystems in the Southwestern United States. There have been a number of studies
pertaining to the edaphic characteristics of eolian deposits
in arid environments of the Southwest (Holiday 1990;
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Simon A. Lei is a Biology and Ecology Instructor, Department of Biology,
WDB, Community College of Southern Nevada, 6375 West Charleston Boulevard, Las Vegas, NV 89146-1193, U.S.A. FAX (702) 257-2914; e-mail:
salei@juno.com
164
Lancaster 1993, 1998; Lei 1998; McFadden and others 1987;
Sharp 1966; Wells and others 1990). Yet, little attention has
been focused on the soil surface and moisture characteristics
of inland dunes in the Mojave Desert.
Despite limited economic and agricultural values, sand
dunes are an important component of the Mojave Desert
landscape, and are ecologically and edaphically interesting
(Lei 1998). Two hypotheses were made prior to data collection: (1) soil moisture attributes would differ among the
three dune sites, between dune and nondune habitats, and
between two geomorphic surfaces (terrace and slope) within
a habitat, and (2) gravimetric soil moisture would be partially influenced by habitat type and soil depth. These two
hypotheses were tested by a combination of field and laboratory measurements of soil moisture attributes in the Kelso,
Ash Meadows, and Death Valley dunes with their adjacent
Larrea tridentata-Ambrosia dumosa (creosote bush-white
bursage) shrublands (nondunes) in the Mojave Desert of
southern Nevada and California.
Methods and Materials ___________
Study Site
Field studies were conducted at three dune systems in
southern Nevada and California (fig. 1) during the summer
of 2001. The Kelso and Death Valley dunes are in southern
California, and the Ash Meadows dunes are in southern
Nevada (fig. 1). Elevations varied considerably, ranging
from 85 m below sea level in Death Valley to 670 m above sea
level in Kelso and Ash Meadows (table 1). Among the three
active dunes, Death Valley has the highest mean monthly
air temperature (fig. 2) and the lowest mean monthly precipitation (fig. 3).
The substrates were deep, unstable, and coarse grained on
all three dunes. Within each dune system, variations in sand
particles were observed. Sand particles, ranging from coarse
to fine in texture, can be distinguished according to their
particle size distribution and mean grain size (Lancaster
1993). Different sand occupies distinct areas of the dune
field, suggesting that dunes represent a stacked sequence of
sand deposition events (Lei 1998).
Atriplex spp. (saltbush) and Larrea tridentata commonly
occur in dune and marginal dune habitats in the Kelso, Ash
Meadows, and Death Valley dunes. Marginal dunes were
located between active dunes and adjacent nondune areas,
with sand more or less stabilized. Abundant vegetation may
eventually cause soil to develop. Larrea plants are typically
larger and taller in dune fields compared to its nondune
USDA Forest Service Proceedings RMRS-P-31. 2004
Soil Moisture Attributes of Three Inland Sand Dunes in the Mojave Desert
Lei
25
N
119°00'W
Amargosa Valley
Death Valley
Twentynine Palms
39°35'N
N E VA D A
2
Mean monthly precipitation (mm)
111°00'W
U TA H
Lake
Mead
1
CALIFORNIA
Colorado River
3
ARIZONA
20
15
10
5
0
1
Salton Sea
100
3
4
5
6
7
8
9
10 11 12
Month
Figure 2—Mean monthly precipitation in
Twentynine Palms, Amargosa Valley, and
Death Valley (local climatological data,
NOAA). Ash Meadows lies within the
Amargosa Valley, while Kelso Dunes lies
approximately 20 km northeast of
Twentynine Palms. Weather data are not
available in Kelso Dunes.
33°00'N
0
2
200
Kilometers
Figure 1—Location of three inland sand dunes in
southern Nevada and California. Dune systems
are numbered as follows: 1 = Ash Meadows, 2 =
Death Valley, and 3 = Kelso.
40
Mean monthly air temperature (°C)
counterpart (Lei 1999). Larrea and Ambrosia are codominant shrub species occurring in the adjacent nondune habitats of the Kelso Dunes and Ash Meadows, while LarreaAtriplex spp. form a common shrub association in Death
Valley. Hymenoclea salsola (cheesebush) and Salsola spp.
(Russian-thistle) are frequently found in dune habitats of
Ash Meadows. Achnatherum hymenoides (Indian ricegrass)
and Astragalus lentiginosus (locoweed) are herbaceous plants
occurring in Kelso Dunes (Lei 1999).
Field and Laboratory Measurements
For each sand dune, 14 transects were established with
each transect containing dune habitats, ecotone (marginal
dune), and adjacent nondune shrublands. These 14 transects
were randomly distributed around the entire dune area.
2
Dune plots, with an area of 100 m , were located at least 200 m
into the dune environments, whereas nondune plots were
located 200 m from the nearest marginal dunes.
At the center of each adjacent nondune plot, one soil
sample was extracted to depths of 5 cm and 20 cm only due
to the presence of caliche layers near the surface. However,
at the center of each dune plot, one soil sample was extracted
at multiple depths—5, 20, 40, 90, and 130 cm. Soils were
30
20
10
Amargosa Valley
Death Valley
Twentynine Palms
0
0
1
2
3
4
5
6
7
8
9
10 11 12
Month
Figure 3—Mean monthly air temperature in
Twentynine Palms, Amargosa Valley, and
Death Valley (local climatological data,
NOAA). Ash Meadows lies within the
Amargosa Valley, while Kelso Dunes lies
approximately 20 km northeast of Twentynine
Palms. Weather data are not available in
Kelso Dunes.
Table 1—Geographic characteristics of three inland sand dunes in the Mojave Desert.
Dune system
Ash Meadows
Death Valley
Kelso
County, State
Latitude
Longitude
Elevation
Nye, NV
Inyo, CA
San Bernadino, CA
--N-36∞20'
36∞35'
34∞55'
--W-116∞15'
116∞00'
115∞40'
-m670
< 85
670
USDA Forest Service Proceedings RMRS-P-31. 2004
165
Lei
passed through a 2-mm sieve to remove large rocks and plant
roots. A soil sieve also separated sand, silt, and clay from
material larger than 2 mm such as gravels and cobbles. Soil
samples were measured for gravimetric moisture. Soils were
oven dried at 105 ∞C for 72 hours. Gravimetric soil moisture
(soil water content) was determined by calculating the
differences between fresh and oven-dried mass divided by
dry mass. To ensure compatibility, air temperatures in the
field were recorded concurrently when soil samples were
collected.
Water infiltration rates were measured using a PVC pipe
(cylinder), 5.5 cm in diameter and 9.5 cm tall. This cylinder
was open at both ends, and was gently tamped into the dune
and nondune soils to a depth of 2 cm to avoid leakage, and
then 50 ml of water was ponded above the core inside of the
cylinder (Brotherson and Rushforth 1983). Infiltration into
the core was measured with a stopwatch as the number of
seconds needed for the ponded water to disappear into the
core (Brotherson and Rushforth 1983).
Approximately 1.5 L of water, serving as artificial rain,
was manually poured through a perforated 13-cm disk, with
perforations evenly spaced on a 0.1-cm grid. The disk was
placed 1.0 m above ground. The total delivery time for the
water to be dispensed on the dune or nondune soil was 60
seconds to create precipitation at a cloudburst level
(Brotherson and Rushforth 1983). A sudden, heavy precipitation is significant due to its impacts on surface-water
runoff and fluvial erosion. Depth of water penetration was
immediately measured and recorded once the water had
soaked into the soil.
Surface-water runoff was measured by recording the
downslope and across-slope spread of water that was artificially rained onto study sites (Brotherson and Rushforth
1983). The area of water spread from these two slope measurements was computed using the formula for the area of
an ellipse using the following equation: (pab), where a and b
are radii of an ellipse (Larson and others 1994). Since surface
water runoff did not form a perfect elliptical shape, measured areas were likely to be overestimates.
Soil movement was assessed by visually quantifying the
amount of soil loss via fluvial erosion during a measured
cloudburst. The following index was used at the spot of this
cloudburst: 1 = no appreciable movement; 2 = moderate
movement—up to 10 percent of soil particles being displaced, and 3 = heavy movement—between 10 and 20 percent of soil particles being displaced (Brotherson and
Rushforth 1983).
Statistical Analyses
One-way Analysis of Variance (ANOVA; Analytical Software 1994) was used to detect significant effects of site
(three dunes), habitat type (dune, marginal dune, and
adjacent shrublands), and soil depth on soil water content.
Tukey and Scheffe’s multiple comparison tests (Analytical
Software 1994) were used to compare site means when
significant effects of habitat type and soil depth were
detected. Pearson’s Correlation (Analytical Software 1994)
was conducted to correlate soil water content with site,
habitat type, and geomorphic surface (terrace and slope).
Path analysis includes a path diagram (fig. 4) with direct
causal pathways whose strengths were indicated by partial
166
Soil Moisture Attributes of Three Inland Sand Dunes in the Mojave Desert
Site
-0.08ns
Soil depth
Habitat type
0.42*
0.77***
Gravimetric soil moisture
Figure 4—Path diagram depicting proposed causal
influences of site, habitat type, and soil depth
(environmental attributes) on gravemtrica soil moisture.
The direct causal effect of each attribute (straight, oneheaded arrow) on soil moisture is the standardized
partial regression coefficient (path coefficient). A short,
unlabeled (residual) arrow is shown to indicate that
additional climatic, edaphic, and geomorphic factors
may play a role in determining the overall soil moisture
content.
regression coefficients (path coefficients), with arrows indicating the direction of causal connection from predictor
(site, habitat type, and soil depth) to response (soil moisture) variables. A short, unlabeled (residual) arrow appeared in the path diagram, indicating that other abiotic
factors would have some influence on the overall gravimetric soil moisture.
Multivariate Analysis of Variance (MANOVA; Analytical
Software 1994) was performed on soil moisture attributes
with site, habitat type, and geomorphic surface as main
effects. P-values equal to or less than 0.05 are reported as
statistically significant.
Results ________________________
Gravimetric soil moisture were significantly different in
site, habitat type, and soil depth (P < 0.01). Dune substrates
had a significantly lower (P < 0.001) water content at the
upper 20 cm than nondune substrates. Soil water content
reached as high as 1.48 percent at 130 cm beneath the
surface in the Ash Meadows dunes during midsummer.
Among the three dune fields, Death Valley dunes consistently exhibited the lowest soil water content at all depths
(table 2). Mean summer precipitation fell below 5 mm (fig. 2),
while mean summer air temperature exceeded 30 ∞C in Death
Valley (fig. 3). In all three dunes, soil water content increased
significantly as soil depth increased (P < 0.001; table 2). Path
analysis revealed significant direct causal effects of soil depth
and habitat type on soil water content (fig. 4). Despite more
extreme monthly and annual weather conditions found in
Death Valley, dune site had little direct effect on soil water
content (table 3). According to path analysis, soil water
content was significantly directly influenced by soil depth and
USDA Forest Service Proceedings RMRS-P-31. 2004
Soil Moisture Attributes of Three Inland Sand Dunes in the Mojave Desert
Lei
Table 2—Soil water content (percentages) at various depths in dune and adjacent nondune (Larrea-Ambrosia) habitats during
July 2001 in the Mojave Desert (n = 14 per habitat in each dune system). Due to the presence of caliche layers near
the soil surface, soil moisture was not examined beyond 20 cm in nondune habitatsa.
Ash Meadows
Dune
Nondune
Soil depth
- - cm - 5
20
40
90
130
0.12aA
.34bA
.64c
.98d
1.48e
0.26aB
.79bB
—
—
—
Death Valley
Dune
Nondune
Dune
Kelso
Nondune
0.07aA
.15bA
.41c
.74d
1.08e
0.15aA
.39bA
.59c
.92d
1.37e
0.33aB
.84bB
—
—
—
0.18aB
.46bB
—
—
—
a
Mean values within each dune system (in rows) followed by different upper case letters are significantly different at P < 0.05. Mean values
within each habitat (in columns) followed by different lower case letters are also statistically significant at P < 0.05.
Table 3—Relationship between three environmental variables and soil water
content. The direct causal effect of each variable is the standardized
partial regression coefficient (path coefficient). Total causal influence
(Pearson’s r-value) sums all direct and indirect pathways.
Variable
Direct causal
Site
Habitat type
Soil depth
a
–0.08
.42b
.70c
Indirect causal
a
–0.09
.23a
.16a
Total causal
–0.17a
.65c
.94c
a
Nonsignificant
Significance levels P < 0.05.
Significance levels P < 0.001.
b
c
habitat type. Soil water content was subject to additional
biotic and abiotic influences, representing by a residual arrow
(fig. 4). Although not quantitatively examined in this study,
other direct and indirect causal effects on soil water content
may include absorption of water by plants, gravity, slope
exposure, topography, macropore, soil compaction, soil particle size, air temperature, precipitation, evaporation, relative humidity, and amount of cloud cover.
Moreover, time for water infiltration and depth of water
penetration increased significantly (P < 0.001; table 4) from
the surrounding Larrea-Ambrosia shrublands to dune fields.
Among the three dunes, Death Valley had a significantly
shorter time for water infiltration and greater depth of
water penetration. Downslope, across slope, and area of
water spread were significantly smaller (P < 0.001; table 4)
in dune than in nondune substrates. Water-born soil movement (fluvial erosion) was significantly greater at slope
than at terrace sites (P < 0.01; table 4) regardless of habitat
types.
Among the soil moisture regimes, significant interaction
(P < 0.001; table 5) was detected between habitat type and
geomorphic surface for area of water spread (surface water
runoff) only. When examining habitat type or geomorphic
surface alone, significant differences (P < 0.05; table 5) were
found in all measured moisture variables. There were also
significant differences (P < 0.05; table 5) in site for water
infiltration and water movement into the soil.
USDA Forest Service Proceedings RMRS-P-31. 2004
Discussion _____________________
The Kelso, Ash Meadows, and Death Valley dunes in the
Mojave Desert of southern Nevada and California were
edaphically distinct in terms of soil moisture status from the
surrounding Larrea-Ambrosia shrublands. In all dunes, soil
water content increased with depth, and decreased from the
surrounding Larrea-Ambrosia shrublands to dune fields.
Because of larger pore size, coarser particles, extremely
high percent sand, and less organic matter, dune soils
provide less moisture retention than nondune soils. In this
study, soil moisture in the top 20 cm was less than 1 percent,
but soil moisture increased significantly as depth increased
in all three dunes during the midsummer season. Death
Valley had significantly lower soil water content than Kelso
Dunes and Ash Meadows, primarily due to a combination of
higher evaporation and summer air temperature, lower
summer precipitation, relative humidity, cloud cover, and
lower elevation. Low moisture availability, usually less than
1 percent at or near the soil surface, is a common characteristic of inland sand dunes in Western America, including the
Eureka, Mojave Desert dunes, and northeastern Colorado
dunes (Bowers 1986; Lei 1998). Water does not store well in
shallow soils, mainly due to rapid evaporation. Conversely,
water is stored well in deep soils and can support perennial
plants with deep roots. The loss of water in deep soils is
primarily by percolation into groundwater, absorption of
plant roots, and transpiration from plant leaves (Prill 1968).
Moisture in dune sand comes from precipitation (Bowers
167
Lei
Soil Moisture Attributes of Three Inland Sand Dunes in the Mojave Desert
Table 4—Moisture characteristics of dune and adjacent nondune (Larrea-Ambrosia shrubland) soils in southern Nevada and
California of the Mojave Desert (n = 14 per habitat in each dune system).
Moisture parameter
Ash Meadows
Dune
Nondune
Death Valley
Dune
Nondune
Kelso
Dune
Nondune
Infiltration rate
Terrace
Slope
- - - - - - - - - - - - - - - - - - - - - - - - - - seconds - - - - - - - - - - - - - - - - - - - - - - - - - - - - 138.7
214.3
121.3
189.4
132.4
218.6
212.3
241.1
201.8
229.7
229.5
256.9
Depth of water penetration
Terrace
Slope
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - cm- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 3.6
3.2
3.7
3.4
3.8
3.5
3.2
2.9
3.4
3.1
3.3
2.9
Downslope spread
Terrace
Slope
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - cm- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 51.7
67.4
45.1
70.8
48.7
75.4
62.1
85.7
53.2
86.4
59.8
89.7
Across slope spread
Terrace
Slope
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - cm- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 36.1
61.2
33.6
59.4
34.4
62.7
44.0
74.2
39.9
68.5
41.9
70.2
Area of spread
Terrace
Slope
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - cm2 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 933.2
2,062.4
757.7
2,012.8
837.6
2,363.8
1366.2
3,179.5
1,061.3
2,959.2
1,252.8
3,148.5
Soil movement
Terrace
Slope
- - - - - - - - - - - - - - - - - - - - - - - - - - - fluvial erosion - - - - - - - - - - - - - - - - - - - - - - - - - 1.4
1.2
1.4
1.2
1.4
1.
1.6
1.4
1.7
1.4
1.7
1.3
Table 5—MANOVA results (P-values) of the effects of habitat type, geomorphic surface, site, and their interactions on six soil
moisture attributes: water infiltration, depth of water penetration, downslope and across-slope water spread, area of
water spread, and water-borne soil movement.
Source
Habitat type (A)
Geomorphic surface (B)
Site (C)
A, B
A, C
B, C
A, B, C
Infiltration
0.0000
.0005
.0005
.1497
.7301
.7118
.9153
Depth
Downslope
Across slope
Area
Soil
0.0000
.0000
.0295
.1372
.6726
.6014
.2372
0.0001
.0137
.7024
.5041
.5942
.9770
.9689
0.0000
.0001
.2750
.3752
.9868
.6710
.7804
0.0000
.0000
.1663
.0028
.6100
.8228
.9396
0.0001
.0006
.7065
.2562
.3743
.7065
.3641
1982). Shortly after a rain, moisture at the soil surface may
be as high as 3 percent (Ramaley 1939). Sandy soils have
large pores, which drain water downward by gravity after
soils are moistened at the surface. Abundant water may be
available in inland sand dunes below the surface layers,
with the top 20 cm of dry sand insulating moister sand below
from air exchange (Bowers 1982; Sharp 1966).
In this study, dune sand had considerably more available
water than its sandy texture would suggest. For instance,
percolating water can reach depths such that it is immune
to evaporation but available to plant roots. Another process is vapor-phase addition of water to sand. This mechanism occurs as dew and mist that condense during the night
on cold sand particles. As heating occurs during the day,
some of this water evaporates into the atmosphere. A significant portion of water vapor, however, moves downward by
diffusion to the cooler sand. The addition of moisture to
sandy desert soils via this mechanism is essential. The
hardpan restricts rooting in nondune sites. Because there
are no root restriction zones in dunes, plants can root to
168
much deeper depths so moisture stored at depth can be
extracted.
Water percolated into dune soils significantly faster than
into nondune soils in this study. The infiltration rate of
South African soils is influenced by many factors, including
organic matter, soil texture, and slope (Dean and Yeaton
1993). Significant interaction was found between habitat
type and geomorphic surface for surface water runoff. Substantially higher water infiltration and greater depth of
water penetration are expected on extremely sandy dune
soils with terrace surface compared to adjacent nondune
soils with slope surface, presumably due to a larger pore size,
higher moisture leaching rate, and lower compaction.
Additionally, soil can be removed by the action of heavy
rains or flash floods. In this study, dune soils exhibited a
significantly lower surface-water runoff compared to nondune
soils because dune soils absorbed more water during and
shortly after a simulated cloudburst. With increased infiltration and reduced surface-water runoff, deeper penetration of water into the soil occurred at terrace than at slope
USDA Forest Service Proceedings RMRS-P-31. 2004
Soil Moisture Attributes of Three Inland Sand Dunes in the Mojave Desert
sites. Nevertheless, a significantly greater fluvial erosion
was detected at slope than at terrace sites in both dune and
nondune soils. Although a short cloudburst did not create
extensive fluvial erosion, some movement of soil particles
occurred at the surfaces of slope sites when water traveled
rapidly downslope by gravity. However, substantially more
surface-water runoff and soil loss via fluvial erosion are
likely to occur in both dune and nondune substrates if
cloudbursts have longer duration, higher frequency, and/or
greater intensity.
Active inland sand dunes occupy only a tiny fraction of the
North America desert landscape. The Kelso, Ash Meadows,
and Death Valley dunes of the Mojave Desert are different
from the surrounding Larrea-Ambrosia shrublands in terms
of soil moisture attributes. Compared to adjacent shrublands,
sand dunes are edaphic islands that have water available at
depth and that are important in shaping unique biogeographic patterns through time. Future research is required
to investigate additional soil attributes, such as chemical
nutrients and biological properties, in inland dunes of North
America and on other continents.
Acknowledgments ______________
I sincerely appreciate Steven Lei, David Valenzuela, and
Shevaun Valenzuela for collecting soil samples and moisture data in the dune fields. Steven Lei assisted with statistical analyses. Scott Keiser provided stimulating discussions, and David Charlet provided helpful comments on
earlier versions of this manuscript. The Department of
Biology at the Community College of Southern Nevada
provided logistical support.
References _____________________
Analytical Software. 1994. Statistix 4.1, an interactive statistical
program for microcomputers. Minneapolis, MN: Analytical Software. 429 p.
Barbour, M. G.; Burk, J. H.; Pitts, W. D. 1999. Terrestrial plant
ecology. 3d ed. Menlo Park, CA: Addison Wesley Longman, Inc.
USDA Forest Service Proceedings RMRS-P-31. 2004
Lei
Bowers, J. E. 1982. The plant ecology of inland dunes in Western
North America. Journal of Arid Environments. 5: 199–220.
Bowers, J. E. 1986. Seasons of the wind. Flagstaff, AZ: Northland
Press. 156 p.
Brotherson, J. D.; Rushforth, S. R. 1983. Influence of cryptogamic
crusts on moisture relationships of soils in Navajo National
Monument, Arizona. Great Basin Naturalist. 43: 73–78.
Brown, J. H. 1972. Species diversity of seed-eating desert rodents in
sand dune habitats. Ecology. 54: 775–787.
Dean, W. R. J.; Yeaton, R. I. 1993. The effects of harvester ant
Messor capensis nest-mounds on the physical and chemical properties of soils in the southern Karoo, South Africa. Journal of Arid
Environments. 25: 249–260.
Holiday, V. T. 1990. Soils and landscape evolution of eolian plains:
the southern High Plains of Texas and New Mexico. Geomorphology. 3: 489–515.
Lancaster, N. 1993. Kelso Dunes. National Geographic Research
and Exploration. 9: 444–459.
Lancaster, N. 1998. The dynamics of star dunes: an example from
the Gran Desierto, Mexico. Sedimentology. 36: 273–289.
Larson, R. E.; Hostetler, R. P.; Edwards, B. H. 1994. Calculus.
Lexington, MA: D.C. Heath and Company.
Lei, S. A. 1998. Soil properties of the Kelso sand dunes in the Mojave
Desert. Southwestern Naturalist. 43: 47–52.
Lei, S. A. 1999. Ecology of psammophytic plants in the Mojave,
Sonoran, and Great Basin Deserts. In: McArthur, E. Durant;
Ostler, W. Kent; Wambolt, Carl L., comps. Proceedings: shrubland ecotones. Proc. RMRS-P-11. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Rocky Mountain Research Station: 212–216.
McFadden, L. D.; Wells, S. G.; Jercinovich, M. J. 1987. Influences of
eolian and pedogenic processes on the origin and evolution of
desert pavements. Geology. 15: 504–508.
National Oceanic and Atmospheric Administration (NOAA). 2001.
Local climatological data: annual summary with comparable
data. Ashville, NC: National Climatic Data Center.
Pavlik, B. M. 1985. Sand dune flora of the Great Basin and Mojave
Deserts of California, Nevada, and Oregon. Madrono. 32: 197–213.
Prill, R. C. 1968. Movement of moisture in the unsaturated zone in
a dune area, southwestern Kansas. U.S. Geological Survey Professional Paper 666D: 1–12.
Ramaley, F. 1939. Sand-hill vegetation of northeastern Colorado.
Ecological Monographs. 9: 1–51.
Sharp, R. P. 1966. Kelso Dunes, Mojave Desert, California. Geological Society of America Bulletin. 77: 1045–1074.
Wells, S. G.; McFadden, L. D.; Schultz, J. D. 1990. Eolian landscape
evolution and soil formation in the Chaco Dune field, southern
Colorado Plateau, New Mexico. Geomorphology. 3: 517–546.
169
Leaf Temperatures of Blackbrush:
Relationships With Air and Soil
Surface Temperatures
Simon A. Lei
Abstract: Comparative diurnal and seasonal leaf temperatures of
blackbrush (Coleogyne ramosissima Torr.) plants were quantitatively examined in the Clark Mountain of southeastern California.
Path analysis revealed that seasonality and diurnal period were
significant positive predictors of air, soil surface, and blackbrush
leaf temperatures. Path analysis also indicated that air temperature was a significant positive predictor of soil surface and blackbrush
leaf temperatures. Blackbrush leaf temperatures consistently remained at or near air temperatures throughout the day and year.
Sunlit leaf temperatures were significantly warmer than shade leaf
temperatures during the summer season only, but not during rest
of the seasons. Soil surface temperatures were significantly warmer
than air and blackbrush leaf temperatures irrespective of seasonality and diurnal period. Significant interactions were detected between seasonality and diurnal period for air and blackbrush leaf
temperatures in the Mojave Desert of southeastern California.
Introduction ____________________
Plant species have adapted to survive various environmental stresses in deserts such as low precipitation, low
relative humidity, low soil moisture and nutrients, intense
solar radiation, and high air and soil temperatures. Solar
radiation on plants growing in warm, arid environments
may cause their death if plant temperatures rise well above
a critical level (Gates and others 1968). Leaf temperatures
exceeding the air temperature by 10 ∞C are common in the
Great Basin Desert of southern Utah (Gates and others
1968). Xerophytes must prevent sunlit leaves from becoming too warm when air temperature exceeds 50 ∞C or more.
Leaf temperatures of many xerophytic species in southern
Utah are very close to air temperatures (Gates and others
1968). Many warm desert plants have small leaves compared to their closely related species living in more mesic
environments. There are ecomorphological reasons why the
temperature of small leaves stay near air temperature.
Large, thin leaves have a greater surface area, absorb heat,
and lose water via transpiration more rapidly compared to
small, thick leaves.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Simon A. Lei is a Biology and Ecology Instructor, Department of Biology,
WDB, Community College of Southern Nevada, 6375 West Charleston Boulevard, Las Vegas, NV 89146-1139, U.S.A. FAX (702) 257-2914; e-mail:
salei@juno.com
170
Temperatures of the sunlit soil surfaces ranged from 12 to
28 ∞C above air temperatures of the Great Basin Desert in
southern Utah (Gates and others 1968). Summer seasons
have the largest daily and annual range of soil surface
temperatures of any season. However, any litter, debris, and
vegetation covering the ground would substantially reduce
the diurnal and annual range of soil surface temperatures in
southern Nevada (Lei and Walker 1997b). Relative to open
substrate temperatures, Suzan and others (1996) reported a
15 ∞C decrease in summer maximum temperatures and an
increase in winter minimum temperatures of 3 ∞C under
shrubs in the Sonoran Desert. During the winter when the
ground is continuously covered with snow, soil surface
temperatures remain at or a few degrees below freezing
because the insulating capacity of the snow traps heat
radiating from the earth, thus protecting plants from extreme cold (Barbour and others 1999).
Leaf temperatures of some desert plant species were
quantified (Gates and others 1968; Gibbs and Patten 1970;
Smith 1978). Yet, possible interactions and correlations
between air and soil surface temperatures in determining
leaf temperatures of blackbrush (Coleogyne ramosissima)
shrubs in southeastern California of the Mojave Desert have
not been documented.
Methods _______________________
Study Site
Field studies were conducted during four 3-month intervals from March through December 2001 at the Clark Mountain (roughly 35∞41’N and 115∞32’W; elevation 1,475 m) near
the Nevada State border. The months of March, June,
September, and December represented the four distinct
seasons. Summers display warm, arid conditions, with maximum air temperatures in the 40 ∞C range. Winter air temperatures, on the contrary, are generally mild and pleasant.
Winter rainfalls are mild and may last up to several days
(Rowlands and others 1977). Monsoonal precipitation occurs
from July through mid-September, as thunderstorms of
high intensity and short duration. Soil erosion, especially
along wash edges, is evidence of the intensity of some of the
thunderstorm activity. The spring and fall seasons are
generally considered most ideal for blackbrush growth and
development, although sharp temperature changes can occur during these months. A relative humidity of 20 percent
or less is common in summer seasons (Lei and Walker
1997a,b). A combination of high air temperatures and high
evaporation, as well as low relative humidity, cloud cover,
USDA Forest Service Proceedings RMRS-P-31. 2004
Leaf Temperatures of Blackbrush Relationships With Air and Soil Surface Temperatures
and low precipitation create a typical arid environment,
with an average annual rainfall of less than 200 mm at
midelevations of the Clark Mountain (Lei, personal observations, 2001).
Blackbrush shrublands often are considered a floristically
simple community, consisting primarily of monospecific
blackbrush stands with other woody species scattered
throughout. Many common herbaceous plants are members
of the Asteraceae, Agavaceae, Brassicaceae, Fabaceae, and
Poaceae families. Blackbrush shrublands share a broad
lower ecotone with creosote bush-white bursage (Larrea
tridentata-Ambrosia dumosa) shrubland, and a broad upper
ecotone with pinyon-Utah juniper (Pinus monophyllaJuniperus osteosperma) woodland (Lei and Walker 1997a,b).
Field Observations
Diurnal and seasonal air temperatures were measured
concurrently with soil surface and blackbrush leaf temperatures to ensure comparability. Air temperatures were measured with a mercury-in-glass thermometer placing at 1.6 m
above ground surface. Soil temperatures were measured
with a metallic soil thermometer placed at the surface in
open substrates and beneath shrub canopies.
Blackbrush shrubs were tagged with a brightly colored
yarn prior to leaf temperature measurements to facilitate
repeated sampling throughout the day. Air, leaf, and soil
surface temperatures were measured at predawn (0600
hours), and then at 3-hour intervals through 1800 hours.
Temperatures of 100 leaves from 50 blackbrush shrubs
were taken with a fine-wire (0.02-mm) copper-constantan
thermocouple inserted into leaf tissues. These 100 leaves
were subdivided evenly between the top and bottom layers
of canopy from the same plants to represent sunlit and
shade leaves, respectively. Individual leaves within each
layer of canopy were randomly selected to avoid biased
sampling.
Statistical Analyses
Two-way Analysis of Variance (ANOVA; Analytical Software 1994) was performed to determine if air temperatures
differed by seasonality (spring and summer) and/or by diurnal period (hour of day). Three-way Analysis of Variance
(ANOVA; Analytical Software 1994) was conducted to detect
significant effects of seasonality, diurnal period, and habitat
type (open substrate and beneath shrub canopy) on soil
temperatures. This three-way ANOVA was also used to
detect significant effects of seasonality, diurnal period, and
canopy position (top and bottom) on leaf temperatures. Path
analysis, expressed as path coefficient or partial regression
analysis (Analytical Software 1994), was performed to determine if diurnal period and seasonality had direct causal
effects on air, soil surface, and blackbrush leaf temperatures. Path analysis was also performed to determine if air
temperature had a causal effect on soil surface and blackbrush
leaf temperatures. Linear regression analysis (Analytical
Software 1994) was used to correlate soil surface temperatures with blackbrush leaf temperatures. Statistical significance was tested at the 5-percent level.
USDA Forest Service Proceedings RMRS-P-31. 2004
Lei
Results ________________________
Air, soil surface, and blackbrush leaf temperatures were
significantly directly affected by seasonality and diurnal
period (P < 0.001; fig. 1). Soil surface and blackbrush leaf
temperatures were also significantly directly influenced by
air temperatures (P < 0.001; fig. 2). Soil surface temperatures were significantly positively correlated with blackbrush
leaf temperature, expressed as a curved, two-headed arrow
(P < 0.001; figs. 1 and 2).
For ease of visualization, air, soil surface, and blackbrush
leaf temperature patterns were extremely similar between
the spring and fall seasons, and only the spring season
values were included graphically. Blackbrush leaf temperatures consistently remained at or near air temperature
throughout the day and year (fig. 3). The air-sunlit leaf
temperature difference in magnitude became greater around
noon hours during the summer season (fig. 3). Sunlit leaves
located at the top of canopies were warmer relative to shade
leaves located deep inside the canopies (fig. 4). Soil surface
temperatures were significantly warmer than air (P < 0.01;
fig. 5) and blackbrush leaf (fig. 3) temperatures irrespective
of diurnal period and seasonality. A similar trend in temperature differences was also observed between air and soil
beneath blackbrush shrub canopy, between air and shade
leaves, as well as between shade leaves and soil beneath
canopy (data not shown).
Soil surface temperatures ranged from slightly below air
temperature in December to nearly 16 ∞C above air temperatures in June (fig. 6). Within a single day, the soil-air
Air temperature
0.94***
0.69***
0.94***
0.96***
Diurnal period
Seasonality
0.94***
0.72***
Plant temperature
0.94***
0.74***
0.92***
Soil temperature
Figure 1—Path diagram depicting hypothesized
relationships between blackbrush leaf temperatures
and various environmental attributes. The direct
causal effect of each attribute (straight, one-headed
arrows) on blackbrush leaf temperature is the
standardized partial regression coefficient (path
coefficient). A short, unlabeled (residual) arrow is
shown to indicate that additional biotic and abiotic
factors may play a role in determining the overall leaf
temperature. Soil temperatures were measured in
open substrates, while leaf temperatures were
measured at the top of the canopy exposed to
prolonged direct sunlight. *** = P values less than
0.001.
171
Lei
Leaf Temperatures of Blackbrush Relationships With Air and Soil Surface Temperatures
4
Spring
Air temperature
0.94***
0.69***
0.79***
0.92***
Diurnal period
Seasonality
0.94***
0.72***
0.94***
Leaf temperature
0.76***
Soil temperature
0.81***
Figure 2—Path diagram depicting hypothesized
relationships between blackbrush leaf temperatures
and various environmental attributes. The direct
causal effect of each attribute (straight, one-headed
arrows) on blackbrush leaf temperature is the
standardized partial regression coefficient (path
coefficient). A short, unlabeled (residual) arrow is
shown to indicate that additional biotic and abiotic
factors may play a role in determining the overall
leaf temperature. Soil temperatures were measured
under shrub canopies. Leaf temperatures were
measured at the bottom of the canopy, and leaves
were shaded throughout much of the day. *** = P
values less than 0.001.
Leaf temperature difference (°C)
Summer
Winter
3
2
1
0
6
18
20
Spring
Summer
Soil-leaf temperature difference (°C)
Air-leaf temperature difference (°C)
15
Figure 4—Temperature difference (n = 100 per
diurnal period per season) between sunlit and shade
leaves in the Clark Mountain of southeastern
California. Shade leaves were located deep inside
the canopy, and leaves were shaded throughout
much of the day.
Spring
Winter
1
0
-1
-2
-3
Summer
Winter
15
10
5
0
6
9
12
15
Hour of day
Figure 3—Temperature difference (n = 100 per
diurnal period per season) between air and sunlit
leaf exposed to prolonged, direct sunlight in the
Clark Mountain of southeastern California.
172
12
Hour of day
3
2
9
18
6
9
12
15
18
Hour of day
Figure 5—Temperature difference (n = 100 per
diurnal period per season) between sunlit soil surface
(open substrate) and sunlit leaves in the Clark
Mountain of southeastern California.
USDA Forest Service Proceedings RMRS-P-31. 2004
Leaf Temperatures of Blackbrush Relationships With Air and Soil Surface Temperatures
Significant interactions were detected between seasonality and diurnal period for air and blackbrush leaf temperatures (table 1). Sunlit leaf temperatures were significantly
warmer (P < 0.05; table 1) relative to shade leaf temperatures. All possible interactions (seasonality, diurnal period,
and habitat type) were found to be significant for soil surface
temperatures (table 1).
20
Soil-air temperature difference (°C)
Spring
Summer
15
Winter
10
Discussion _____________________
5
0
-5
6
9
12
15
18
Hour of day
Figure 6—Temperature difference (n = 100 per diurnal
period per season) between air and sunlit soil surface
(open substrate) in the Clark Mountain of southeastern
California.
temperature differences in magnitude became greater around
midafternoon hours irrespective of seasonality (fig. 6). Soil
surface temperatures in open substrate were significantly
warmer (P < 0.01) than soil temperature under shrub canopies during the dry summer season, but were not significantly different during the snowy winter season (fig. 7).
20
Spring
Summer
Soil temperature difference (°C)
Lei
15
Winter
10
5
0
-5
6
9
12
15
18
Hour of day
Figure 7—Temperature difference (n = 100 per diurnal period per
season) between sunlit soil surface (open substrates) and soil beneath
shrub canopies in the Clark Mountain of southeastern California.
USDA Forest Service Proceedings RMRS-P-31. 2004
Diurnal and seasonal temperature profiles were used to
compare differences in magnitude of air, soil surface, and
blackbrush leaves. Moreover, habitat type and canopy position are included as an additional variable for measuring soil
surface temperatures and blackbrush leaf temperatures,
respectively. The connections among environmental variables are represented in path diagrams by two types of
arrows: a straight, one-headed arrow signifies a causal
relationship between two variables, and a curved, twoheaded arrow signifies a simple correlation between them
(Loehlin 1998).
In this study, path analysis indicated that seasonality and
diurnal period were significant positive predictors of air, soil
surface, and blackbrush leaf temperatures. This analysis
also revealed that air temperature was a significant positive
predictor of soil surface and blackbrush leaf temperatures.
Comparing to other seasons, the air-soil surface temperature difference in magnitude became substantially greater
around midafternoon hours in open substrates during the
dry summer season. Among all measured temperatures, the
hottest temperatures were observed on soil surfaces in open
substrates, exceeding both air and blackbrush leaf temperatures by as much as 15.8 ∞C in the summer season. Temperatures at soil surface in open substrates were significantly
warmer, and experienced more seasonal extremes than
temperatures at soil surface beneath shrub canopies in
southern Nevada (Lei 1995; Lei and Walker 1997b) because
the bare soil is a good heat absorber and it heats up much
more rapidly.
According to the path diagram, a short, unlabeled (residual) arrow is pointed at air temperature, implying that
variation in air temperature is subject to additional influences besides seasonality and diurnal period. Air and soil
temperatures vary in a fairly parallel manner. Since air
temperature is chiefly dependent on radiation and conduction from and to the ground, it follows that the diurnal and
annual courses of the air temperatures and the ground
temperatures should be similar (Shanks 1956).
A residual arrow is also pointed at soil surface temperatures in the path diagram. Soil surface temperatures are
largely determined by a combination of air temperature, the
presence of soil moisture and resulting evaporation, the
presence of litter, debris, and/or vegetation cover, as well as
the radiation regime incident upon the soil surface. During
winter seasons, the presence of moisture at the soil surface
and the resulting evaporation tends to give a low and
uniform soil temperature (Shanks 1956). A persistent snow
cover maintains the soil temperatures at or near 0 ∞C regardless of the air temperature, and even insulates the soil
sufficiently to allow warmth from below to cause a slight rise
in soil temperature. Conversely, a lack of snow cover allows
173
Lei
Leaf Temperatures of Blackbrush Relationships With Air and Soil Surface Temperatures
Table 1—Summary of two-way ANOVA showing the effects of seasonality, dirunal period, and their interactions on air
temperatures. Summary of three-way ANOVA showing the effects of seasonality, diurnal period, habitat type or
canopy position, and their interactions on soil surface and blackbrush leaf temperatures, respectively. df = 1 for
seasonality, habitat type, and canopy position; df = 4 for diurnal period.
Variable
Seasonality (A)
Diurnal period (B)
Habitat type (C)
Canopy position (C)
A, B
B, C
A, C
A, B, C
F
Air temperature
P
7,512.80
347.71
—
—
13.17
—
—
—
0.0000
.0000
—
—
0.0008
—
—
—
the soil temperature to average lower than the freezing point
under frigid winter conditions (Shanks 1956).
Soil surface and blackbrush leaf temperature were significantly correlated with, and have a significant direct effect
on, each other in this study. A dense vegetation cover or any
covering on the soil considerably lessens the diurnal and
annual surface temperature ranges. Due to the presence of
dense blackbrush shrub canopies, soil surface temperatures
under prolonged shade consistently remained at or near the
air temperatures, with relatively little direct radiation reaching the ground. A large amount of solar radiation is absorbed
by the vegetation rather than by the soil (Suzan and others
1996). In this study, soil temperatures beneath shrub canopies were 2.2 to 15.7 ∞C cooler than soil temperatures in
adjacent open substrates during the summer month, but
were only 0.9 to 2.4 ∞C warmer during the winter month.
Shade decreases soil surface temperatures by 13.5 C to 15 ∞C
in summer seasons, but increases soil temperatures by 1.2 to
3 ∞C in winter seasons in the Sonoran Desert (Franco and
Nobel 1989; Suzan and others 1996). Shade is normally
considered advantageous to seedling establishment in deserts
of southeastern California because it moderates temperature extremes and water loss in microhabitats beneath
shrub canopies (Walker and others 2001). Hence, shade
creates cooler microhabitat temperatures and higher moisture regimes compared to adjacent open substrates.
Direct heating of the soil surface by solar radiation is
important in considering the total heat load in the environment (Gibbs and Patten 1970). Heat is transferred from the
soil to the plants by reradiation, convective air currents, and
conduction by direct physical contact with the plants. Heat
is also conducted from the soil surface to lower portions of
plants where roots and soil exchange heat (Gibbs and Patten
1970). The tendency for canopies of blackbrush to occur near
the ground surfaces allows the leaves to approach high
temperature extremes. A combination of high leaf temperatures, low soil moisture, and low plant water content may
play a vital role in determining the lower elevational limit of
blackbursh in southeastern California of the Mojave Desert.
In this study, leaf temperatures of blackbrush consistently remained at or near air temperatures throughout the
day and year because the leaves are small and dissipate heat
primarily by convection. The dimensions of blackbrush leaves
ranged from 0.5 to 1.4 cm in length, and were less than 1 cm
174
F
Soil temperature
P
14,668.39
775.93
655.16
—
72.74
12.33
76.39
3.07
0.0000
.0000
.0000
29.88
0.0000
.0000
.0000
0.0415
Leaf temperature
F
P
14,423.44
560.71
—
.0000
24.41
.12
1.98
.51
0.0000
.0000
—
—
0.0000
.97
.18
.73
in width. Temperatures of a leaf measuring 1 by 1 cm or less
remain close to air temperatures (Gates and others 1968).
They (1968) also discovered that temperatures of small
leaves in shrubs and trees (Artemisia, Ephedra, Juniperus,
and so forth) were within 3 ∞C of air temperatures in southern Utah, which is in agreement with this study.
Significant interactions were observed between seasonality and diurnal period for leaf temperatures. When evaluating individual variables alone, significant differences were
observed in seasonality, diurnal period, and canopy position
for blackbrush leaf temperatures. The coolest leaves were
found within the leaf canopy, and were shaded from direct
sunlight throughout much of the day. Early morning and
late afternoon sunlight produces relatively less heat stress
in plant tissues than midday sunlight in southern Nevada
(Brittingham and Walker 2000). Reduced leaf temperature
decreases heat damage to tissues, enhances stomatal functioning, maintains protein and membrane integrity, and
lessens photoinhibition in summer dry seasons ( Brittingham
and Walker 2000; Singla and others 1997). Shading also
decreases leaf temperature, thereby reducing water vapor
concentration within intercellular air spaces, and reducing
water vapor concentration gradient between intercellular
air spaces and ambient air (Brittingham and Walker 2000).
In this study, leaf temperatures ranged from 2.8 ∞C below
to 1.2 ∞C above air temperatures depending on seasonality,
diurnal period, and canopy position, as well as on the
intensity, duration, and frequency of direct sunlight. Sun
leaves were considered those that were fully exposed to
sunlight for most of the day. Conversely, shade leaves were
located deeply inside the leaf canopy, mostly shaded during
the day. Shade leaves ranged from 0.5 to 2.3 ∞C cooler than
sun leaves of the same plants in this study.
Ecological Implications __________
The diurnal and seasonal fluctuations of air temperatures
largely determined the fluctuations of soil surface and
blackbrush leaf temperatures. Small leaves found in many
desert plant species are important morphological adaptations in the regulation of leaf temperature. Due to high
transpiration rates and high stomatal conductance, small
leaves may act to reduce leaf temperatures to more optimal
photosynthetic levels, particularly for rapid growth periods
USDA Forest Service Proceedings RMRS-P-31. 2004
Leaf Temperatures of Blackbrush Relationships With Air and Soil Surface Temperatures
when water is abundant (Smith 1978). Without ample water
supply and high transpiration rates, a larger leaf size that
orients horizontally, exposed to prolonged sunlight and
without pubescent layers, can contribute to warmer leaf
temperatures, often well above air temperatures. Blackbrush
is a typical microphyllous desert shrub species, so it is not
surprising that leaf temperatures stay close to air temperatures. This phenomenon has been shown many times for
microphyllous desert species. Thus, unusually high leaf
temperatures for extensive periods can cause long-term or
even permanent (irreversible) plant tissue damage in frequently stressful desert environments.
Acknowledgments ______________
I gratefully acknowledge Steven Lei for collecting air, soil
surface, and plant leaf temperature data in the field and for
assisting with statistical analyses. David Charlet provided
helpful comments on earlier versions of this manuscript.
The Department of Biology at the Community College of
Southern Nevada (CCSN) provided logistical support.
References _____________________
Analytical Software. 1994. Statistix 4.1, an interactive statistical
program for microcomputers. St. Paul, MN: Analytical Software.
Barbour, M. D.; Burks, J. H.; Pitts, W. D. 1999. Terrestrial plant
ecology. Menlo Park, CA: Addison Wesley Longman, Inc.
Brittingham, S.; Walker, L. R. 2000. Facilitation of Yucca brevifolia
recruitment by Mojave Desert shrubs. Western North American
Naturalist. 60: 374–383.
USDA Forest Service Proceedings RMRS-P-31. 2004
Lei
Franco, A. C.; Nobel, P. S. 1989. Effects of nurse plants on the microhabitat and growth of cacti. Journal of Ecology. 77: 870–886.
Gates, D. M.; Alderfer, R.; Taylor, E. 1968. Leaf temperatures of
desert plants. Science. 159: 994–995.
Gibbs, J. G.; Patten, D. T. 1970. Plant temperatures and heat flux
in a Sonoran Desert ecosystem. Oecologia. 5: 165–184.
Gibson, A. C. 1996. Structure-function relations of warm desert
plants. Berlin, Germany: Springer.
Lei, S. A. 1995. A gradient analysis of Coleogyne communities in
southern Nevada. Las Vegas: University of Nevada. Thesis.
Lei, S. A.; Walker, L. R. 1997a. Classification and ordination of
Coleogyne communities in southern Nevada. Great Basin Naturalist. 57: 155–162.
Lei, S. A.; Walker, L. R. 1997b. Biotic and abiotic factors influencing
the distribution of Coleogyne communities in southern Nevada.
Great Basin Naturalist. 57: 163–171.
Loehlin, J. C. 1998. Latent variable models: an introduction to
factor, path and structural analysis. Mahwah, NJ: Larrence
Erlbaum Associates.
Rowlands, P. G.; Johnson, H.; Ritter, E.; Endo, A. 1977. The Mojave
Desert. In: Barbour M.; Major, J., eds. Terrestrial vegetation of
California. New York: John Wiley and Sons, Inc.
Shanks, R. E. 1956. Altitudinal and microclimatic relationships of
soil temperature under natural vegetation. Ecology. 37: 1–7.
Singla, S. L.; Pareek, A.; Grover, A. 1997. Cellular responses to high
temperature. In: Prasad, M. N. V., ed. Plant Physiology. New
York: John Wiley and Sons, Inc.: 101–127.
Smith, W. K. 1978. Temperatures of desert plants: another perspective on the adaptability of leaf size. Science. 201: 614–616.
Suzan, H.; Nabhan, G. P.; Patten, D. T. 1996. The importance of
Olneya tesota as a nurse plant in the Sonoran Desert. Journal of
Vegetation Science. 7: 635–644.
Walker, L. R.; Thompson, D. B.; Landau, F. H. 2001. Experimental
manipulations of fertile islands and nurse plant effects in the
Mojave Desert, USA. Western North American Naturalist. 61:
25–35.
175
Impacts of Livestock Exclusion
From Wyoming Big Sagebrush
Communities
Jennifer M. Muscha
Ann L. Hild
Larry C. Munn
Peter D. Stahl
Abstract: Fenced exclosures are used by range scientists to determine the effects of livestock grazing on vegetation and soils. In the
1950s and 1960s numerous rangeland exclosures were established
on previously grazed areas in Wyoming big sagebrush steppe.
Current understanding suggests shrublands may not return to the
pristine condition with livestock exclusion. In the summers of 2001
and 2002 we examined nine 30- to 45-year-old exclosures in Natrona,
Fremont, Sweetwater, Bighorn, and Washakie Counties in Wyoming
to compare vegetative cover and presence of biological soil crusts
inside and outside exclosures. Bare soil was greater outside exclosures at three sites, and litter cover was greater inside the
exclosure at three of the nine sites. Biological soil crust cover was
greater inside only one of nine exclosures. Although shrub cover
increased from original levels (data collected at time of exclosure
establishment) both inside and outside exclosures at all sites, shrub
cover was greater inside than outside at only one exclosure site. Our
results are mixed with respect to impacts of livestock exclusion from
Wyoming big sagebrush steppe, partially because original sagebrush levels were low in most exclosures, and because grazing use
outside exclosures differed among the nine sites.
Introduction ____________________
Native rangeland protection is becoming increasingly
important, as managers attempt to enhance and conserve
biodiversity. Because a majority of our rangelands are under
public ownership, management choices become controversial. Grazing of domestic livestock has been suggested as
beneficial to semiarid rangelands in the Western United
States (Vavra and others 1994), while others propose permanent removal of grazing by domestic livestock (Donahue
1999; Fleischner 1994; and others). The effects of grazing
removal from already grazed systems in the West are not
well documented.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Jennifer M. Muscha is a Graduate Student at the University of Wyoming.
Ann L. Hild, Larry C. Munn, and Peter D. Stahl are Associate Professor,
Professor, and Assistant Professor, Department of Renewable Resources,
University of Wyoming, Laramie, WY 82071, U.S.A., Fax: (307) 766-6403, e-mail:
annhild@uwyo.edu
176
Previous research on exclosures conducted in arid and
semiarid regions document different vegetative responses to
the exclusion of grazing. The most evident difference between grazed and ungrazed areas is greater density of
aboveground vegetation inside the exclosure (Costello and
Turner 1941). Although range managers once assumed that
removal of grazing pressure from rangelands improved
rangeland condition (Dyksterhuis 1949), several exclosure
studies in arid environments demonstrate that recovery of
degraded rangelands can be slow to nonexistent (Holechek
and Stephenson 1983; Robertson 1971; West and others
1984). Other studies in arid and semiarid rangelands examining removal of livestock from an area report no change in
vegetation (Johnson-Barnard 1995; West and others 1984),
an increase in herbaceous cover but not species richness
(Anderson and Holte 1981), or an increase in shrub cover
after removal of grazing (Brady and others 1989; Lang
1973), resulting in a more degraded state. In sagebrush
systems, noticeable response of vegetation after grazing
removal may not occur for many decades, and removal of
grazing can increase density and cover of big sagebrush
(Anderson and Holte 1981; Bock and others 1984; Lang
1973). If the grazing-free period has been short (less than 20
years; Brady and others 1989), shrub canopy may remain
unchanged or increase greatly (sagebrush ground cover
increased 293 percent inside and 30 percent outside an
exclosure in northeastern Wyoming; Lang 1973).
Removal of livestock grazing from rangelands can be
evaluated by using fenced grazing exclosures to compare
vegetation inside with an adjacent grazed area. Long-term
exclosures can reveal slow vegetative trends; however, several limitations of exclosures have been noted. Vegetative
conditions between the grazed and protected area may have
differed at the time of exclosure establishment (Costello and
Turner 1941). The exclosure itself may act as a barrier to wind
movement and affect soil and moisture deposition, altering
temperature and humidity (Daubenmire 1940). Because
exclosures were often constructed where vegetative change
due to grazing had already occurred, these sites may not
reflect the response of pristine vegetation. However, these
sites do reflect the response of grazed vegetation to cessation
of grazing by domestic livestock. Because range scientists
are finding the Dyksterhuis (1949) view of range succession
is not accurate where woody species dominate, Archer (1994)
has distinguished that shrub-driven succession may deviate
from grass-driven succession. Consequently, long-term
USDA Forest Service Proceedings RMRS-P-31. 2004
Impacts of Livestock Exclusion From Wyoming Sagebrush Communities
Muscha, Hild, Munn, and Stahl
exclosures in shrubland ecosystems can clearly demonstrate
the impacts of grazing removal in much of the semiarid west.
Finally, because biotic crusts are highly susceptible to soil
surface disturbance such as trampling or offroad vehicles
(Belnap and Gardner 1993), and their biomass and biotic
activity are concentrated within 3 mm of the soil surface
(Garcia-Pichel and Belnap 1996), they are usually influenced by domestic livestock use. Although biotic crusts are
often more developed inside Great Basin exclosures
(Kaltenecker and others 1999), little is known of their
presence and richness in Wyoming shrublands. We selected
nine 30- to 45-year-old rangeland exclosures within Wyoming sagebrush steppe and examined shrub and biological
soil crust presence to better understand impacts of grazing
cessation.
Rankin Basin
(1965)
Worland Cattle (1966)
Big Trails (1962)
Lander Ant
(1957)
Between 1959 and 1965 approximately 100 rangeland
exclosures were constructed in Wyoming for a cooperative
study by the University of Wyoming, the Wyoming Agricultural Experiment Station Cooperative program, and the
Bureau of Land Management (Fisser 1967, 1968). Studies of
vegetational change by use of 0.6-m (2-ft) by 6.1-m (20-ft)
and 1.2-m (4-ft) by 1.2-m (4-ft) plots were initiated in 1959
and 1960 (Fisser 1968). The study of vegetation change by
use of permanent 30.3-m (100-ft) transects of 20 plots, each
0.3-m (1-ft) by 0.3-m (1-ft) square was initiated in 1965
(Fisser 1967). Some of these exclosures were sampled
annually until the early 1970s, and all sampling using the
original methods ceased after the early 1980s. A small
number of these exclosures have been extensively studied
since establishment (Garland 1972; Gdara 1977; JohnsonBarnard 1995; Uhlich 1982).
Site Selections and Description
Nine exclosure study sites were selected within Wyoming
big sagebrush (Artemisia tridentata Nutt. ssp. wyomingensis
Beetle and Young) in Natrona, Fremont, Sweetwater, Bighorn, and Washakie Counties in Wyoming (fig. 1). Sites were
selected to have comparable slope and elevation with minimal disturbance (no breaks in fences). Ages of exclosures
ranged from 32 to 46 years (table 1) and ranged in size from
2.5 to 12.5 ha (1 to 5 acres). Dominant grass species found at
exclosure sites were western wheatgrass (Agropyron smithii
Rydb.), Sandberg bluegrass (Poa secunda Presl), bluebunch
wheatgrass (Agropyron spicatum (Pursh) Scribn. & Smith),
junegrass (Koeleria macrantha (Ledeb.) Schult.), needle and
thread (Stipa comata Trin. & Rupr.), blue grama (Bouteloua
gracilis (H.B.K.) Lag. ex Steud.), and Indian ricegrass
(Oryzopsis hymenoides (R. & S.) Ricker) (table 1). Donlin was
the only exclosure dominated by threadleaf sedge (Carex
filifolia Nutt.). Cheatgrass (Bromus tectorum L.) was found
at five of the nine sites: Poison Spider, Big Trails, Powder
Rim D, Worland Cattle, and Sand Creek.
Soil surface texture (Blake and Hartge 1986) was analyzed on soil samples retrieved from inside and outside
exclosures. Grazing history at each exclosure site was
USDA Forest Service Proceedings RMRS-P-31. 2004
Donlin (1958)
Poison Spider
(1965)
Red Wash #2
(1958)
Powder Rim D (1970)
Materials and Methods ___________
Exclosures in Wyoming
Sand Creek (1970)
Figure 1—Locations of nine range exclosures
sampled in Wyoming in 2001 and 2002 (year of
establishment is indicated in parentheses).
maintained at the Bureau of Land Management Regional
offices. Current grazing use and annual precipitation is
reported in table 1. Seven exclosures in our study had sandy
loam soil textures (table 1). The Poison Spider exclosure had
15 percent more sand content inside the exclosure than
outside. Poison Spider and Big Trails have the greatest
annual precipitation of the exclosures examined. Donlin and
Powder Rim D have comparable annual precipitation and
soil textures. Sand Creek and Worland Cattle also have
comparable annual precipitation and soil texture, but are
drier than Donlin and Powder Rim D sites. Rankin Basin
and Red Wash #2 are the driest sites in our study because of
low precipitation, and their soils have greater clay content
than the other seven exclosures (table 1). Most exclosure
sites are grazed in spring or in all seasons on a rotational
basis. Rankin Basin site is different from other exclosures in
its light grazing use only in winter.
Methods
In 2001 and 2002 we placed vegetative line transects 20 m
in length at each exclosure. The number of transects (three
to four inside and outside each exclosure) at each site was
determined by the size of the exclosure. Basal and canopy
cover was recorded on each transect. Transects were evenly
spaced inside the exclosure and permanently marked on
both ends with a metal stake. Outside the exclosure, transects
were placed the same distance from the fence line as the
inside transects. Basal and cover data at each site was
analyzed using a two-group t-test (alpha 0.05) to compare
grazed to grazing removal treatments. We compared our
vegetative data to original records collected by Fisser at the
time of exclosure construction. Because Fisser’s initial
samples included only a single transect in each exclosure
and position (no replicates), we did not conduct analysis of
variance on initial data sets. Some differences in vegetative
cover between initial samples and our data may be partially
177
Muscha, Hild, Munn, and Stahl
Impacts of Livestock Exclusion From Wyoming Sagebrush Communities
Table 1—Site characteristics of nine Wyoming shrubland exclosures sampled in 2001 and 2002.
Exclosure
name
Established
Elevation
Annual
precipitation
Big Trails
year
1962
m
1,515
mm
348
Poison Spider
1965
1,758
Lander Ant
1957
Donlin
Dominant grass/
grasslike species
(in order of abundance)
Soil
texture
Domestic
animal
Months
grazed
Stocking
rate
Acres
per
AUM
3.0
Sandberg bluegrass,
western wheatgrass,
bluebunch wheatgrass,
cheatgrass
Sandy
loam
Cattle,
horses
Nov. to Jan.,
first year
April to June,
second year
Rested
third year
303
Cheatgrass, Sandberg
bluegrass, western
wheatgrass, junegrass
Sandy
loam
Cattle,
sheep
March to June
5.5
1,621
221
Needle and thread,
blue grama, Sandberg
bluegrass, western
wheatgrass
Sandy
loam
Cattle
March to April
Nov. to Feb.
10.0
1958
1,879
285
Threadleaf sedge,
needle and thread
Sandy
loam
Cattle
April-June
Sept. to Oct.
Powder Rim D
1970
1,939
277
Needle and thread,
Sandberg bluegrass
Sandy
loam
Cattle,
sheep
Deferred
rotation, all
Worland Cattle
1970
1,300
194
Needle and thread,
cheatgrass, western
wheatgrass
Sandy
loam
Cattle,
sheep
March to May,
Oct. to Dec.a
9.0
Sand Creek
1966
1,333
194
Needle and thread,
Sandberg bluegrass
Sandy
loam
Cattle,
sheep
March to May,
Oct. to Dec.a
11.9
Rankin Basin
1958
1,932
172
Blue grama,
needle and thread
Sandy
clay loam
Cattle
Oct. to May
18.0
Red Wash #2
1965
1,576
258
Needle and thread,
Indian ricegrass,
bluebunch wheatgrass
Sandy
clay loam
Cattle
Dec. to April
10.0
a
6.1
10.0
Grazing use highly varied in this exclosure since established.
attributed to differences in sampling methods. However,
because our data include only a single observer (Muscha)
this effect is uniform across the nine exclosure sites.
Results ________________________
Exclosure sites represent environments with varieties of
plant available moisture because of precipitation and soil
disparities between sites (table 1). Consequently our tables
and figures present the sites roughly ordered from most
mesic (Big Trails and Poison Spider) to most xeric (Red Wash
#2) based on precipitation and soil texture differences.
Basal cover differed between inside and outside positions
at four of the nine exclosure study sites (P < 0.05). At Donlin,
Big Trails, and Poison Spider, bare soil was greater, and
litter cover was less outside than inside exclosures (fig. 2). In
two of these sites (Poison Spider and Big Trails) litter was
the dominant cover inside exclosures (76 percent and 64
percent, respectively), while at all other sites bare soil was
the dominant cover inside exclosures. Litter in the Poison
Spider exclosure was primarily due to invasion of cheatgrass
178
at this site (fig. 3). Cheatgrass was also present at the Big
Trails site, but absent from Lander Ant. Cheatgrass cover
was greater inside Poison Spider and Big Trails exclosures.
Powder Rim D exclosure was unique because it was the only
site where cover of biological soil crusts differed between the
two positions and was greater inside the exclosure. At all
sites, biological soil crust cover was less than 10 percent
regardless of position.
Vegetative canopy cover differed between positions (inside and outside) at Big Trails, Poison Spider, Powder Rim D,
and Red Wash #2 sites (fig. 3). Native grass cover was
greater inside the exclosure at Powder Rim D (P < 0.05); at
Red Wash #2 and Poison Spider, native grass cover was
greater outside the exclosures.
Only Poison Spider and Rankin Basin exclosures had
initial shrub canopy cover more than 10 percent at the time
of exclosure construction (figs. 4 and 5). In general, shrub
cover increased from time of construction to present at all
exclosure sites, regardless of position, and overall increases
were greater inside exclosures than outside. In the most
recent samples, shrub cover differed between positions (t-test,
<0.05) only at Poison Spider (fig. 5).
USDA Forest Service Proceedings RMRS-P-31. 2004
Impacts of Livestock Exclusion From Wyoming Sagebrush Communities
Muscha, Hild, Munn, and Stahl
Percent ground cover
100
80
*
60
40
20
0
In
Out
Big Trails
In
Out
Poison Spider
In
Out
Lander Ant
In
Out
Donlin
In
Out
In
Out
In
Out
Powder Rim D Worland Cattle Sand Creek
Soil crust
Litter
In
Out
Rankin Basin
In
Out
Red Wash #2
Bare soil
Figure 2—Soil cover (crusts, litter, and bare ground) in two positions (inside and outside grazing exclosures) at nine
sagebrush study sites in Wyoming (within a site and cover type asterisks delineate significance at P < 0.05, t-test).
Percent canopy cover
100
90
80
70
60
50
40
30
20
10
0
In
Out
Big Trails
In
Out
Poison Spider
In
Out
Lander Ant
Shrub
In
Out
Donlin
Native grass
In
Out
In
Out
In
Out
Powder Rim D Worland Cattle Sand Creek
Cheatgrass
In
Out
In
Out
Rankin Basin Red Wash #2
Forbs and succulents
Figure 3—Canopy cover (shrub, native grass, cheatgrass, and forbs and succulents) in two positions (inside and
outside grazing exclosures) at nine sagebrush study sites in Wyoming (within a site and cover type asterisks
delineate significance at P < 0.05, t-test).
Discussion _____________________
Effects of Grazing Removal
The range of plant available moisture represented by the
nine exclosure sites offered a variety of responses to removal
of domestic stock from sagebrush steppe systems. Overall, at
more xeric sites there were few differences inside and outside exclosures. Shrub cover increased both inside and outside exclosures as they aged, regardless of precipitation and
soil differences. The Red Wash #2 site is unique in having
more native grass cover outside the exclosure. This difference
is likely due to a low stocking rate and use in winter months,
allowing for favorable grass regrowth prior to our sampling in
June. Other xeric sites were used in the spring, prior to
sampling. In general, the most xeric sites have shown limited
response to grazing removal so far. Moderately moist sites
USDA Forest Service Proceedings RMRS-P-31. 2004
such as Powder Rim D seem to reflect more response to
grazing removal, and the greatest response is seen in the
most mesic sites (Big Trails and Poison Spider). The Lander
Ant site may be unique in lack of response; however, this site
differs from Big Trails and Poison Spider in its use only by
cattle, lower stocking rate, and slightly lower precipitation.
Additionally, cheatgrass is a large part of the vegetative
cover in the Big Trails and Poison Spider sites and is absent
from Lander Ant. The difference between positions at these
two exclosures demonstrates the potential for transitions to
shrub-driven succession as suggested by Archer (1994) and
others. The unique interactions of available moisture, grazing animal, and season of use are impossible to separate in
this study but provide important insight relative to the
mechanisms involved in sagebrush increase. Additionally,
none of the nine exclosures have been burned since establishment, and so generalizations relative to cheatgrass
179
Muscha, Hild, Munn, and Stahl
Impacts of Livestock Exclusion From Wyoming Sagebrush Communities
Shrub cover (percent)
Inside
Outside
Big Trails
Lander Ant
40
40
30
30
20
20
10
10
0
0
1966
1978
2002
1959
Donlin
40
30
30
20
20
10
10
0
0
1980
1967
2001
40
40
30
30
20
20
10
10
0
0
1978
1987
2001
Sand Creek
Worland Cattle
1966
2001
Powder Rim D
40
1965
1980
1971
2002
Rankin Basin
1978
2002
Red Wash #2
40
40
30
30
20
20
10
10
0
0
1965
1978
2002
1961
1981
2001
Figure 4—Shrub cover change at eight Wyoming exclosure sites from year of
establishment to present (2001 or 2002 sampling dates). Order based on total
canopy cover inside exclosures, soils, and precipitation (Poison Spider shown
in fig. 5).
180
USDA Forest Service Proceedings RMRS-P-31. 2004
Impacts of Livestock Exclusion From Wyoming Sagebrush Communities
shrub canopy to increase. In most cases, shrub canopy cover
increased comparably with or without domestic grazing
livestock use. Explanation for these differential responses to
grazing exclusion requires further examination.
Percent shrub cover
Poison Spider
A
40
30
B
20
10
0
1965
1980
2001
Year
Inside
Outside
Figure 5—Shrub cover in Poison Spider
exclosure from time of establishment to 2001
(within a year, columns with the same letter
do not differ P > 0.05, t-test).
dynamics in Wyoming sagebrush steppe are speculative in
the absence of fire.
Biological Soil Crusts
Biological soil crusts have received increased attention in
arid land research, and our research has found that biological soil crusts are a visible presence in some shrubland
ecosystems in Wyoming. Although we have no records of the
cover of biotic crusts at time of exclosure establishment, we
can only assume that they were similar across the sites prior
to the time of fence construction. Although total cover of
biological soil crusts was significantly greater inside only
one exclosure (Powder Rim D; fig. 2), our data indicate a
general trend of higher cover inside all exclosures. Additionally, crusts remain to be detailed for morphological characteristics or richness as suggested by Rosentreter (this proceedings). Several studies have shown biological soil crusts
are negatively impacted by and highly susceptible to livestock trampling (Anderson and others 1982; Beymer and
Klopatek 1992; Jeffries and Klopatek 1987; and others).
Cover of biological soil crusts was less than 10 percent inside
and outside exclosures at all sites even after more than 40
years of domestic livestock removal and do not appear to
reflect stocking rate differences among our sites.
Conclusions ____________________
The small differences between the inside and outside
positions at our nine exclosure sites indicate that some
shrubland systems will not necessarily improve to a more
productive system after 40 years of livestock exclusion. Our
data demonstrate shrub cover will not always increase
where stock are excluded at a greater rate than with livestock present. It is clear that the release of grazing pressure
does not decrease shrub canopy; in some cases it may allow
USDA Forest Service Proceedings RMRS-P-31. 2004
Muscha, Hild, Munn, and Stahl
References _____________________
Anderson, D. C.; Harper, K. T.; Holmgren, K. T. 1982. Factors
influencing development of cryptogamic crusts in Utah deserts.
Journal of Range Management. 35: 355–359.
Anderson, J. E.; Holte, Karl E. 1981. Vegetation development over
25 years without grazing on sagebrush-dominated rangeland in
southeastern Idaho. Journal of Range Management. 34: 25–29.
Archer, S. 1994. Woody plant encroachment into southwestern
grasslands and savannas: rates, patterns and proximate causes.
In: Vavra, M.; Laycock, W. A.; Pieper, R. D. Ecological implications of livestock herbivory in the West. Denver, CO: Society for
Range Management.
Belnap, J.; Gardner, J. S. 1993. Soils microstructure in soils of the
Colorado Plateau: the role of cyanobacterium Microcoleus
vaginatus. Great Basin Naturalist. 53: 40–47.
Beymer, J. D.; Klopatek, J. M. 1992. Effects of grazing on cryptogamic crusts in pinyon-juniper woodlands in Grand Canyon
National Park. American Midland Naturalist. 127: 139–148.
Blake, G. R.; Hartge, K. H. 1986. Particle-size analysis. In: Klute,
Arnold, ed. Methods of soil analysis. Part 1. Physical and
mineralogical methods. Madison, WI: American Society of
Agronomy, Inc.
Bock, C. E.; Bock, J. H.; Kenney, W. R.; Hawthorne, V. M. 1984.
Responses of birds, rodents, and vegetation to livestock exclosure
in a semidesert grassland site. Journal of Range Management.
37: 239–242.
Brady, W. W.; Stromberg, M. R.; Aldon, E. F.; Bohnam, C. D.; Henry,
S. H. 1989. Response of a semidesert grassland to 16 years of rest
from grazing. Journal of Range Management. 42: 284–288.
Costello, D. F.; Turner, G. T. 1941. Vegetation changes following
exclusion of livestock from grazed ranges. Journal of Forestry. 39:
310–315.
Daubenmire, R. F. 1940. Exclosure technique in ecology. Ecology.
21: 514–515.
Donahue, D. 1999. The western range revisited: removing livestock
from public land to conserve native biodiversity. Norman: University of Oklahoma Press. 388 p.
Dyksterhuis, E. J. 1949. Condition and management of rangeland
based on quantitative ecology. Journal of Range Management. 2:
104–115.
Fisser, H. G. 1967. Exclosure studies with transects of permanent
plots. 1965–1966 results. Science Rep. No. 69. Wyoming Agriculture Experiment Station.
Fisser, H. G. 1968. Long term exclosure studies with 2 x 20 ft and
4 x 4 ft permanent plots. 1959–1964 results. Science Rep. No. 131.
Wyoming Agriculture Experiment Station.
Fleischner, T. L. 1994. Ecological costs of livestock grazing in
Western North America. Conservation Biology. 8: 629–644.
Garcia-Pichel, F.; Belnap, J. 1996. The microenvironments and
microscale productivity of cyanobacterial desert crusts. Journal
of Phycology. 32: 774–782.
Garland, C. B. 1972. Ecology of arid land vegetation in western
Wyoming. Laramie: University of Wyoming. Thesis.
Gdara, A. O. 1977. Interpretation of long term vegetation data from
permanent quadrats on semiarid shrublands of western Wyoming. Laramie: University of Wyoming. Thesis.
Holechek, J. L.; Stephenson, T. 1983. Comparison of big sagebrush
vegetation in north central New Mexico under moderately grazed
and grazing excluded conditions. Journal of Range Management.
36: 455–456.
Jeffries, D. L.; Klopatek, J. M. 1987. Effects of grazing on the
vegetation of the blackbrush association. Journal of Range Management. 40: 390–392.
Johnson-Barnard, J. 1995. Long-term vegetation changes at eight
exclosures in the arid semi-desert of southwestern Wyoming.
Laramie: University of Wyoming. Dissertation.
181
Muscha, Hild, Munn, and Stahl
Kaltenecker, J. H.; Wicklow-Howard, M.; Rosentreter, R. 1999.
Biological soil crusts in three sagebrush communities recovering
from a century of livestock trampling. In: McArthur, E. D.; Ostler,
W. K.; Wambolt, C. L., comps. Proceedings: shrubland ecotones.
Proc. RMRS-P-11. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 222–226.
Lang, R. 1973. Vegetation changes between 1943 and 1965 on the
shortgrass plains of Wyoming. Journal of Range Management.
26: 407–409.
Robertson, J. H. 1971. Changes on a sagebrush-grass range in
Nevada ungrazed for 30 years. Journal of Range Management.
24: 397–400.
182
Impacts of Livestock Exclusion From Wyoming Sagebrush Communities
Uhlich. J. W. 1982. Vegetation dynamics of non-grazed and grazed
sagebrush grassland communities in the Big Horn Basin of
Wyoming. Laramie: University of Wyoming. Thesis.
Vavra, M.; Laycock, W. A.; Pieper, R. D., eds. 1994. Ecological
implications of livestock herbivory in the West. Denver, CO:
Society for Range Mangement.
West N. E.; Provenza, F. D.; Johnson, P. S.; Owens, M. K. 1984.
Vegetation change after 13 years of livestock grazing exclusion on
sagebrush semidesert in west central Utah. 37(3): 262–264.
USDA Forest Service Proceedings RMRS-P-31. 2004
Species Richness Inside and
Outside Long-Term Exclosures
W. A. Laycock
D. L. Bartos
K. D. Klement
Abstract: Recent environmental literature contains claims that
livestock grazing has caused reduction in species diversity on
Western rangelands. Data of species richness (number of species)
is presented from inside and outside 24 long-term exclosures in
Montana, Utah, and Wyoming. For the average of all exclosures
there was no difference between species richness inside and
outside the exclosures. Overlap of species (those found both inside
and outside of the exclosures) was high (69 to 72 percent), and the
species found either inside only or outside only did not reveal any
trend related to grazing or successional status. The areas inside
and outside eight exclosures in Wyoming were sampled in 2
successive years. Growing season moisture was above normal the
first year and far below normal the second. Species richness
sampled was more than 40 percent lower in the second year than
in the first. The rather small difference in species richness inside
and outside all of the exclosures indicates that neither grazing nor
lack of grazing has much long-term influence on species richness.
However, growing season moisture can greatly influence how
many species are encountered when sampling an area.
Introduction ____________________
Some conservation biology and environmental literature
contains claims that livestock grazing has caused and
continues to cause reduction in species diversity on Western rangelands: Livestock have greatly lowered plant diversity in most of the West (Jacobs 1991). One of the
ecological costs of livestock grazing is the “reduction of
species richness” (Fleischner 1994). Livestock grazing is
the most insidious and pervasive threat to biodiversity on
rangelands (Noss and Cooperrider 1994). Livestock grazing has altered and diminished the presettlement diversity
of native fauna and flora on many Western rangelands
(Donahue 1999). Ehrlich (1990), Wuerthner (1990), and
Noss (1994) made similar statements. Wuerthner (1994)
lumped farming and pastoralism (grazing) together and
suggested that agriculture is the most serious threat to
biodiversity in the Western United States. In most of these
publications, only anecdotal information was presented to
back up the claims that grazing reduces species diversity.
Authors have not always clearly defined terms (like biodiversity, diversity, and species diversity) used in their publications. Because of this, some definition of terms may be
useful at this point. West (1993) defined “biological diversity” (or “biodiversity”) as:
...the variety of life and its processes, including the variety of
living organisms, the genetic differences among them, the
communities, ecosystems and landscapes in which they occur,
plus the interactions of these components.
This definition specifically includes diversity at the genetic,
species, community, or ecosystem and landscape or regional
level. In a later publication, West (1995) suggested that the
culture and indigenous knowledge of local people engaged in
sustainable lifestyles should also be added to the definition
of biodiversity.
Species diversity is usually expressed in one of two ways:
richness, which is the number of species present in an area;
and evenness or equitability, which is a measure of the
abundance of each species present (West 1993). This paper
deals with only one element of diversity, that is, that of plant
species richness.
This paper compares species richness (number of plant
species) inside and outside 24 long-term exclosures in Utah,
Wyoming, and Montana. Only species richness is reported
because the data were collected by different methods in each
area and, in some cases, species lists were pooled from more
than one sampling method, and species evenness or other
indices of diversity could not be calculated. Species richness
inside and outside the eight exclosures in Wyoming also was
compared over 2 years with substantially different amounts
of growing season precipitation. This illustrated how such
precipitation can influence the number of species encountered in sampling.
Methods _______________________
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
W. A. Laycock is a Rangeland Consultant, 3415 Alta Vista Drive, Laramie,
WY 82072, U.S.A.; e-mail: blaycock@wyoming.com. D. L. Bartos is a Range
Scientist, Rocky Mountain Research Station, USDA Forest Service, Logan,
UT. K. D. Klement is a Rangeland Support Scientist, USDA Agricultural
Research Service, Miles City, MT.
USDA Forest Service Proceedings RMRS-P-31. 2004
The central Utah study included eight exclosures with
quaking aspen (Populus tremuloides) vegetation communities. One exclosure was established in 1934, and the other
seven were established between 1947 and 1974. The exclosures ranged in size from 0.3 to 1.0 ha. Cover of understory
2
vegetation by species was estimated on 0.1-m quadrats in
1995 or 1996. Overstory tree species were not included in the
species richness numbers unless the tree species occurred in
183
Laycock, Bartos, and Klement
Species Richness Inside and Outside Long-Term Exclosures
the understory. The overall comparisons for these exclosures were published by Kay and Bartos (2000), but the
species richness data for each individual exclosure were
obtained from file data.
In southwestern Wyoming, eight exclosures in sagebrush
(Artemisia tridentata ssp. Wyomingensis) and/or salt desert
shrub vegetation communities were studied. One was established in 1940, and the other seven exclosures were established between 1958 and 1963. The exclosures were either
0.4 or 0.8 ha in size. For shrubs, cover by species was sampled
with line and belt transects. Herbaceous species were sampled
2
with a point frame and density counts on 1.2-m quadrats.
Species lists from the various methods were pooled to calculate species richness. The same sampling methods were
repeated by the same observer in 1993 and 1994. The species
richness information for these exclosures were obtained
from Johnson-Barnard (1995).
The southwestern Montana study included eight highelevation exclosures: three in tall forb, three in open conifer,
and two in mountain big sagebrush (Artemisia tridentata
ssp. vaseyana) vegetation communities. Seven exclosures
2
are 15.0 m (0.02 ha) and were established in 1978. One of
the open conifer exclosures is 0.09 ha in size and was
established in 1960. For the open conifer type, only understory species were sampled, and the mature overstory tree
species are not included in the number of species. Herbage
production (biomass) by species was estimated in 1994 on
2
2
1.0-m quadrats for sagebrush and 0.5-m quadrats for other
vegetation types. The species richness data for these
exclosures were obtained from Klement (1997) and from file
data.
All 24 exclosures had been protected from livestock grazing all or a majority of the time since establishment. The
areas outside the exclosures had been grazed by livestock
during the same seasonal period in each respective location:
by cattle and/or sheep during the summer in Utah; by cattle
and/or sheep either in summer or year round in Wyoming;
and by sheep only during the summer in Montana. All areas
were available for wildlife use. Plant species nomenclature
follows that was used in each individual study.
Results and Discussion __________
Aspen Exclosures (Utah)
The understory plant communities in the aspen exclosures were not especially species-rich and had the fewest
number of species of all the vegetation types studied (table
1). The average number of species detected by sampling was
about the same both inside (9.3) and outside (9.1) the
exclosures. Two exclosures had the most species found
inside (12) and two different exclosures had the most species
found outside (11). Three exclosures had more species inside
the exclosure than outside, and four exclosures had more
species outside the exclosure than inside. One exclosure had
the same numbers of species inside and outside (table 2).
A total of 12 species were found inside one or more of the
eight exclosures but not outside. Twelve species were found
outside one or more of the eight exclosures but not inside
(table 3). Bromus ciliatus and Sitanion hystrix were found
inside but not outside at two exclosures, but these species
were found both inside and outside at four other exclosures.
All other species were found either inside only or outside
only at just one exclosure.
The overlap of species, that is, the number of species found
both inside and outside, averaged 7.5 species (69 percent) for
Table 1—Average number of plant species inside and outside aspen, shrub, and high-elevation exclosures.
Vegetation type
Location
Life form
Aspen (Utah)
Shrub (Wyoming)
Tall forb
High elevation (Montana)
Sagebrush
Open conifer
All exclosures
Inside exclosure
Shrub
Grass
Forb
Total
- - - - - - - - - - - - - - - - - - - - - - - -Average number of species - - - - - - - - - -- - - - - - - - - - - - 1.9
7.0
0
2.5
0.3
0.8
4.4
4.4
6.7
5.5
8.7
7.1
3.0
2.6
27.7
21.5
22.0
24.0
9.3
14.0
34.4
29.5
31.0
31.9
Outside exclosure
Shrub
Grass
Forb
Total
2.6
3.5
3.0
9.1
7.2
4.9
2.4
14.5
0
7.3
29.0
36.3
2.0
5.5
24.5
32.0
0
12.3
25.7
38.0
.5
8.8
26.6
35.9
Table 2—Number of exclosures with different numbers of species inside or outside for each vegetation type.
Vegetation type
Location of different
numbers of species
More species inside than outside
More species outside than inside
Same number of species inside and outside
184
Aspen
Shrub
Tall forb
High elevation
Sagebrush
Open conifer
All exclosures
- - - - - - - - - - - - - - - - - - - - - - - - - Number of exclosures - - - - - - - - - - -- - - - - - - - - - - - 3
3
1
0
0
1
4
4
2
2
3
7
1
1
0
0
0
0
USDA Forest Service Proceedings RMRS-P-31. 2004
Species Richness Inside and Outside Long-Term Exclosures
Table 3—Species found either inside or outside, but not
both, for the eight aspen exclosures in Utah.
Species inside but not outside
Shrubs
Chrysothamnus viscidiflorus
Grasses and grasslike
Bromus ciliatus
Festuca idahoensis
Festuca thuberi
Sitanion hystrix
Stipa comata
Stipa lettermanii
Forbs
Aquilegia coerula
Galium spp.
Potentilla gracilis
Total species
Inside but not outside = 12
Laycock, Bartos, and Klement
mule deer populations were low. The species richness data
did not reflect these composition changes in the vegetation
in response to use by cattle and deer.
The aspen exclosures tended to be on drier sites, which
may partially explain the low number of plant species
sampled. A more detailed study of two aspen exclosures in
Utah was published by Mueggler and Bartos (1977). One of
these exclosures, Big Flat, was also sampled in 1995–1996
and was included in the Kay and Bartos (2000) study.
Shrub Exclosures (Wyoming)
Species outside but not inside
Shrubs
Artemisia tripartita
Juniperus comunis
Juniperus osteosperma
Pushia tridentata
Rosa woodsii
Symphoricarops oreophilus
Grasses
Muhlenbergia wrightii
Poa pratensis
Forbs
Antennaria microphylla
Castilleja linariafolia
Taraxacum officinale
Total species
Outside but not inside = 12
the eight exclosures (table 4). The lowest overlap was 58
percent, and the highest was 90 percent.
Species richness inside and outside the exclosures was
almost identical and did not reveal any grazing effect.
However, Kay and Bartos (2000) indicated that understory
species composition of the aspen stands studied had been
affected by ungulate herbivory. In the area available to
mule deer but excluding livestock, tall forbs and shrubs
were reduced, while grasses and unpalatable forbs increased. Cattle grazing tended to reduce grass cover. Both
livestock and deer browsing affected aspen regeneration;
aspen regenerated successfully in the livestock exclosure
and combined use plots outside the exclosures only when
The average number of species sampled was similar inside
(14.0) and outside (14.5) the exclosures (table 1). Approximately half of the species were shrubs. The largest number
of species found both inside and outside was 18 at different
exclosures. More species were found inside the exclosure
than outside at three exclosures, more species were found
outside than inside at four exclosures, and the same number
of species occurred inside and outside one exclosure (table 2).
When all eight exclosures were combined, 17 species were
found inside but not outside a given exclosure, and 19 species
were found outside but not inside. Most species were found
only inside or only outside the exclosure at just one site, but
one species was found inside only at two sites, one species
was found outside only at two sites, and two species were
found outside only at three sites.
One forb and two wheatgrass species were found inside only
at one site and outside only at a different site. Five other forb
species were found inside only at one exclosure, and one other
forb species was found outside only at three exclosures. Two
species of shrubs were found inside the exclosure only at one
site and outside the exclosures only at two different sites. Six
other species of shrubs were found inside one or more exclosures
only but not outside, while seven completely different shrub
species were found outside one or more exclosures only but not
inside. Other than two wheatgrasses, no other grasses were
found inside only, but five additional grass species were found
outside only at one exclosure. The average overlap of species
found both inside and outside was 11.9 species or 72 percent
(table 4). The highest overlap of species occurring inside and
outside the same exclosure was 83 percent, and the lowest
overlap was 56 percent.
The number of species found either inside the exclosures
only or outside only did not give any indication that grazing
had a different effect on the vegetation of these shrub
exclosures than 35 years of protection from grazing. JohnsonBarnard (1995) found no observable effect of grazing on
cover, density, production, or species composition of shrub or
Table 4—Plant species overlap—average number of species per exclosure in common (present both inside and outside), found inside exclosures
only and found outside exclosures only for the three vegetation types.
Vegetation type
Aspen (Utah)
Species in common
Inside but not outside
Outside but not inside
Shrub (Wyoming)
Tall forb
High elevation (Montana)
Sagebrush
Open conifer
All exclosures
- - - - - - - - - - - - - - - - - - - - - - - - - - Number of species (percent) - - - - - - - - - - -- - - - - - - - - - - - - - - 7.5 (69)
11.9 (72)
30.7 (77)
26.0 (73)
26.7 (63)
28.0 (70)
1.8 (16)
2.1 (13)
3.7 (9)
3.5 (10)
4.3 (10)
3.8 (10)
1.6 (15)
2.6 (15)
5.7 (14)
6.0 (17)
11.3 (27)
7.9 (20)
USDA Forest Service Proceedings RMRS-P-31. 2004
185
Laycock, Bartos, and Klement
Species Richness Inside and Outside Long-Term Exclosures
herbaceous species for these exclosures. The only consistent
difference between inside and outside was that shrub height
was significantly (P = 0.05) greater inside all of the exclosures.
High-Elevation Exclosures (Montana)
The high-elevation exclosures in Montana had greater
species richness than the Utah and Wyoming vegetation
types sampled, averaging more than 30 species both inside
and outside the exclosures (table 1). A great many of the
species of grasses and forbs were the same in all three
vegetation types for the high-elevation exclosures. The percentage of the total number of species made up by forbs was
68 to 81 percent for all of the vegetation types (table 1)
(Klement 1997).
In the tall forb type, the highest number of species was 39
inside two exclosures and 42 outside one of the same
exclosures. In the mountain big sagebrush type, one exclosure
had 30 species inside and 32 species outside. This mountain
big sagebrush type had more than twice the total number of
species as the lower elevation Wyoming big sagebrush
exclosures sampled in Wyoming.
In the open conifer type, the highest number of species was
41 outside and 38 inside the same exclosure. For all vegetation types, more species occurred outside than inside at
seven of the exclosures (table 2), with an average of four
more species outside. One of the tall forb exclosures had
three more species inside than outside.
In spite of the rather large numbers of species found only
inside or only outside the exclosures, the overlap (species
found both inside and outside a given exclosure) was rather
high for all of the high-elevation exclosures, ranging from an
average 26.0 to 30.7 species in common, with an average
overlap of 70 percent (table 4).
Even though more species generally were found outside of
the exclosures than inside, species richness did not reveal
any major differences in vegetation composition between the
inside and the outside the exclosures in all three vegetation
types. This agrees with Klement (1997) who, based on the
biomass data, concluded that light summer grazing by sheep
does not appear to be a factor in vegetation composition in
these high-elevation vegetation types.
Effect of Growing Season Precipitation on
Species Richness
The areas inside and outside the eight sagebrush and salt
desert shrub exclosures in Wyoming were sampled by the
same methods by the same observer in both 1993 and 1994.
The 1994 data were summarized above. Late spring and
early summer precipitation (May through July) from the
nearest weather station (Kemmerer, WY) indicated that
growing season precipitation in the 2 years were quite
different. The long-term average for that period was 7.39 cm.
Growing season precipitation was 31 percent above this
average (9.70 cm) in 1993 and 73 percent below this average
(2.77 cm) in 1994.
As might be expected, the number of shrub species sampled
inside and outside were the same in 1993 and 1994. However
the total number of herbaceous species sampled was 41
percent lower in 1994 than in 1993 (table 5) and the changes
were similar both inside and outside the exclosures. The
largest reduction in number of species was for forbs. The
number of annual forb species sampled was 69 percent lower
inside and 84 percent lower outside in 1994 than in 1993.
The number of perennial forb species sampled was 48 percent lower inside and 45 percent lower outside. The reduction in number of grass species sampled in 1994 was smaller,
30 percent lower inside and 28 percent lower outside.
The species that were not found in 1994 presumably did
not disappear from the community. The annual forb species
may simply not have germinated in the dry conditions of
1994. The perennial forbs may have grown earlier when soil
moisture was present but dried and were absent or unrecognizable at the time of sampling. The grass species absent in
1994 may have had so little growth that they were not
recognizable at the time of sampling. The important point is
that sampling in a year with low growing season precipitation can result in a lower number of species being detected
than would be detected in a more favorable year. The 22
species that were present in 1993, but absent in 1994, inside
exclosures at one or more sites and outside exclosures at one
or more sites are listed in table 6.
Similar changes in numbers of species sampled as a
result of differences in precipitation have been reported.
On northern mixed grass prairie in North Dakota, Biondini
and others (1998) reported an increase in the number of
forb species sampled from 14 in 1988 at the end of a
prolonged drought to 36 in 1995 after several years of above
average precipitation. These changes occurred in pastures
grazed moderately and heavily by cattle and also in exclosures. At the Cedar Creek Natural History Area in Minnesota, Tilman and El Haddi (1992) reported that the local
species richness of four different grassland fields decreased
an average of 37 percent during a 1988 drought. However,
despite the return to more normal precipitation during the
Table 5—Sagebrush and salt desert shrub exclosures in Wyoming. Changes in richness of
herbaceous species in 2 years with different precipitation.
Life form
186
Inside
1993
Outside
Inside
1994
Outside
1993–1994 change
Inside
Outside
Grass
Perennial forb
Annual forb
- - - - - - - - - - - - Average number of species - - - - - - - - - - - - - - - 6.3
6.8
4.4
4.9
–1.9
–1.9
4.0
3.8
2.1
2.1
–1.9
–1.7
1.6
1.9
.5
.3
–1.1
–1.6
Total
11.9
12.5
7.0
7.3
–4.9
–5.2
USDA Forest Service Proceedings RMRS-P-31. 2004
Species Richness Inside and Outside Long-Term Exclosures
Laycock, Bartos, and Klement
Table 6—Sagebrush and Salt Desert Shrub Exclosures in
Wyoming. Herbaceous species found in 1993 but not
in 1994 inside exclosures at one or more different
sites and outside exclosures at one or more different
sites.
Number of exclosures
Inside
Outside
Grasses
Agropyron dasystachyum
Agropyron smithii
Agropyron spicatum
Oryzopsis hymenoides
Poa spp.
Poa canby
Poa fendleriana
Poa secunda
Sitanion hystrix
Stipa comata
5
2
3
1
3
1
1
1
3
1
6
1
1
1
1
1
1
1
4
2
Perennial forbs
Allium textile
Antennaria dimorpha
Astragalus convallarius
Cryptantha fendleri
Eriogonum flavum
Eriogonum ovalifolium
Phlox hoodii
Phlox longifolia
1
2
1
1
1
1
3
3
4
2
1
1
1
1
2
3
Annual forbs
Astragalus spp.
Descurainia pinnata
Halogeton glomeratus
Lappula redowskii
2
1
1
2
3
1
1
4
next 2 years, there was no significant recovery of species
richness in the permanent plots. They concluded that
“environmentally extreme conditions can limit species richness by causing the local extinction of rare species.” However, they did not present long-term data to indicate that
rare species had, indeed, become extinct in their permanent plots.
Conclusions ____________________
Because the sampling methods employed in this study
were not primarily intended to compile a complete list of all
species in the area, not every species present inside or
outside each exclosure would have been encountered. To
compile a comprehensive species list would have required
sampling at different times during the growing season and
additional searches for rare species outside the sample
quadrats or lines. However, because the vegetation at each
exclosure site was sampled by the same person, using the
same methods and at the same time both inside and outside
of each exclosure, comparisons of species richness between
the two areas should be valid.
Species richness, by itself, may or may not be a good
indicator of the effects of grazing on vegetation, depending
on the situation. For the shrub areas in Wyoming and the
high-elevation areas in Montana, the lack of difference in
USDA Forest Service Proceedings RMRS-P-31. 2004
species richness inside and outside the exclosures agreed
with the conclusions from analysis of cover or biomass data,
that is, the vegetation inside and outside the exclosures was
not different. However, for the aspen areas in Utah, the
cover data revealed some differences caused by grazing or
browsing by livestock or deer that the species richness
comparison did not reveal.
The rather large impact that weather, mainly growing
season precipitation, can have on the number of species
encountered in a sample was shown by comparing samples
taken in 2 successive years in Wyoming. The number of
herbaceous species sampled in the dry summer of 1994 was
more than 40 percent lower than the number sampled in the
relatively wet summer of 1993. The majority of the species
that were present in 1993 but not in 1994 were perennials that
had not disappeared from the area. They either did not grow
in 1994, or they may have been indistinguishable during
sampling because they were too small or unidentifiable.
Collins (1987) reported that areas of tall grass prairie both
burned and moderately grazed by cattle had the highest
plant species diversity, while areas burned but not grazed
had the lowest diversity. On shortgrass steppe vegetation in
Colorado, Hart (2001) found that plant species diversity and
evenness were greatest in lightly and moderately grazed
pastures, least in exclosures, and intermediate in heavily
grazed pastures. These pastures had been grazed by cattle
at the same intensities for 55 years, and the exclosures had
been protected from grazing during the same period.
Laycock (1994) suggested that either no grazing or very
heavy grazing may reduce species diversity, but moderate
grazing probably does not diminish diversity and may enhance diversity at both the species and landscape levels by
promoting patchiness of vegetation.
Oba and others (2000) reviewed literature from African
rangelands that suggested that grazing at moderate intensities on Mediterranean, sub-Saharan, and other subtropical rangelands increased plant species diversity and that, on
Sahelan rangelands, neither continuous grazing nor longterm protection had a lasting effect on plant species diversity. Oba and others (2000) concluded that:
Thus, the evidence suggests that the lack of grazing may
degrade rangelands in sub-Saharan Africa. Long-term deferral of grazing results in either no significant changes in
species composition or in lowered diversity caused by the
disappearance of herbivory-adapted species. Therefore, if the
removal of ungulate grazing results in the loss of principal
species from the plant community, then this ecosystem may
be considered to have been degraded in the absence of grazing.
The results of the present study agreed with published
information summarized above that moderate livestock grazing does not cause plant species richness to decline. Thus,
the claims that livestock grazing has reduced plant species
richness on Western rangelands appear to be without merit
at least for the vegetation types studied on these 24 sites.
Studies need to be conducted on many more rangeland
vegetation types to determine the effects of livestock grazing
on species richness or other elements of biodiversity. Previous studies have the potential to provide such information,
but species richness or other measures of species diversity
usually cannot be calculated from published results of grazing or exclosure studies because journals seldom allow the
187
Laycock, Bartos, and Klement
publication of complete species lists and often lump minor
species as “Other.” Complete lists can be obtained from file
data for previous studies or from complete species lists in
unpublished theses or dissertations.
References _____________________
Biondini, M. E.; Patton, B. D.; Nyren, P. E. 1998. Grazing intensity
and ecosystem processes in a northern mixed-prairie, USA.
Ecological Applications. 8: 469–479.
Collins, S. L. 1987. Interaction of disturbances in tallgrass prairie:
a field experiment. Ecology. 68: 1243–1250.
Donahue, D. 1999. The Western range revisited: removing livestock
from public lands to conserve native biodiversity. Norman: University of Oklahoma Press. 388 p.
Ehrlich, P. R. 1990, Habitats in crisis: why should we care about loss
of species: Forest Ecology and Management. 35: 5–11.
Fleischner, T. L. 1994. Ecological costs of livestock grazing in
Western North America. Conservation Biology. 8: 629–644.
Hart, R. H. 2001. Plant biodiversity on shortgrass steppe after 55
years of zero, light, moderate or heavy cattle grazing. Plant
Ecology. 155: 111–118.
Jacobs, L. 1991. Waste of the West: public lands ranching. Tucson,
AZ: Lynn Jacobs. 602 p.
Johnson-Barnard, J. 1995. Long term vegetation changes at eight
exclosures in the arid semi-desert of southwestern Wyoming.
Laramie: University of Wyoming, Department of Rangeland
Ecology and Watershed Management. Dissertation.
188
Species Richness Inside and Outside Long-Term Exclosures
Kay, C. E.; Bartos, D. L. 2000. Ungulate herbivory on aspen
communities on Utah aspen: assessment of long-term exclosures.
Journal of Range Management. 53: 145–153.
Klement, K. D. 1997. Condition and trend analysis on subalpine
rangeland grazed by sheep in the Centennial Mountains, Montana. Laramie: University of Wyoming, Department of Rangeland Ecology and Watershed Management. Thesis.
Laycock, W. A. 1994. Implications grazing vs. no grazing on today’s
rangelands. In: Vavra, M.; Laycock, W. A.; Pieper, R. D., eds.
Ecological implications of livestock herbivory in the West. Denver, CO: Society for Range Management: 250–288.
Noss, R. F. 1994. Cows and conservation biology. Conservation
Biology. 8: 613–616.
Noss., R. F.; Cooperrider, A. Y. 1994. Saving nature’s legacy:
protecting and restoring biodiversity. Washington, DC: Island
Press.
Oba, G.; Stenseth, N. C.; Lusigi, W. J. 2000. New perspectives on
sustainable grazing management in arid zones of sub-Saharan
Africa. BioScience. 50: 35–51.
Tilman, D.; Haddi, A. El. 1992. Drought and biodiversity in grasslands. Oecologia. 89: 257–264.
West, N. E. 1993. Biodiversity of rangelands. Journal of Range
Management. 46: 2–13.
West, N. E. 1995. Biodiversity on rangelands: definitions and
values. In: West, N. E., ed. Biodiversity on rangelands. Logan:
Utah State University, College of Natural Resources. Natural
Resources and Environmental Issues. Volume IV: 1–4.
Wuerthner, G. 1990. The price is wrong. Sierra. (September): 38–43.
Wuerthner, G. 1994. Subdivision versus agriculture. Conservation
Biology. 8: 905–908.
USDA Forest Service Proceedings RMRS-P-31. 2004
Vegetation Management for
Sagebrush-Associated Wildlife
Species
J. Kent McAdoo
Sherman R. Swanson
Brad W. Schultz
Peter F. Brussard
Abstract: It is obvious that the diverse array of wildlife species
using sagebrush habitats has a similarly wide range of habitat
requirements. Vegetation management for biological diversity on a
landscape scale should take these diverse habitat requirements into
consideration. Management for any one species may or may not
provide the habitat requirements for other species. Creating a
mosaic of habitats with multiple-aged stands of sagebrush and
varying degrees of herbaceous and shrub cover would provide the
diverse vertical and horizontal vegetation composition and structure required by diverse wildlife species.
Introduction ____________________
The sagebrush (Artemisia spp.)-dominated area of the
West encompasses approximately 155.5 million acres (Paige
and Ritter 1999). More specifically, there are two major
regions of sagebrush dominance (West 1983a,b): (1) the
“sagebrush steppe” that covers the northern portion of the
Intermountain Region from eastern Washington and northern Nevada to the western two-thirds of Wyoming and
northwest Colorado, and (2) the “sagebrush semidesert,” the
drier Great Basin sagebrush region that includes most of
Nevada, parts of Utah and northern Arizona, and some
areas of southwestern Colorado and northern New Mexico.
The sagebrush semidesert is significantly drier and warmer
than the sagebrush steppe and occurs between the sagebrush steppe and the drier salt-desert shrub region. Although sagebrush communities have undergone much change
in modern history, the boundaries of sagebrush distribution
have remained fairly constant.
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McAruthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
J. Kent McAdoo is Area Rangeland Resources Specialist, University of
Nevada Cooperative Extension, 1500 College Parkway, Elko, NV 89081,
U.S.A., phone: (775) 738-7291; e-mail: mcadook@unce.unr.edu. Sherman R.
Swanson is Professor of Range Science, University of Nevada-Reno, Reno,
NV; Brad W. Schultz is Extension Educator, University of Nevada Cooperative Extension, Winnemucca, NV. Peter F. Brussard is Professor of Biology,
University of Nevada-Reno, Reno, NV.
Authors’ note: Portions of this paper have been previously presented and
included as an expanded abstract in the non-peer-reviewed proceedings of the
Restoration and Management of Sagebrush/Grass Communities Workshop,
November 4–8, 2002, Elko, NV.
USDA Forest Service Proceedings RMRS-P-31. 2004
Vale (1974) examined 29 historic journals and diaries
from the early 19th century and concluded that presettlement
vegetation in much of the Intermountain West was visually
dominated by shrubs, with much of the area covered by thick
brush stands. However, according to Miller and Eddleman
(2001), fire influences caused plant composition to vary from
dominant stands of sagebrush to grasslands. The authors
speculated that much of the sagebrush steppe ecosystem
during pre-settlement times was composed of open shrub
stands with a substantial herbaceous cover component.
Extreme weather conditions and insect outbreaks also affected the historic patterns of vegetation composition in
sagebrush habitats.
Sagebrush-grass communities within these sagebrush
regions vary markedly (Miller and Eddleman 2001; Tisdale
and Hironaka 1981; West 1983a,b), but to one degree or
another provide food, thermal cover, escape routes, and
rearing sites for a variety of vertebrate wildlife species
(McAdoo and Klebenow 1979). Some of these species inhabit
sagebrush habitats year round, while others use them only
seasonally or occasionally. Species that require sagebrush
for some part of their life cycle are “sagebrush obligates.”
According to Paige and Ritter (1999), at least eight vertebrate species are considered to be sagebrush obligates: sage
sparrow (Amphispiza belli), Brewer’s sparrow (Spizella
breweri), sage thrasher (Oreoscoptes montanus), sage grouse
(Centrocercus urophasianus), pygmy rabbit (Sylvilagus
idahoensis), sagebrush vole (Lagurus curtatus), pronghorn
antelope (Antilocapra americana), and sagebrush lizard
(Sceloporus graciosus). However, the latter species is also
found in greasewood (Sarcobatus spp.) habitats (McAdoo
and Klebenow 1979) and may not be a true “obligate.” In
parts of their range, gray flycatchers (Empidonax wrightii)
and least chipmunks (Eutamias minimus) may also be
considered sagebrush obligates. A number of other species
have a broader amplitude of habitat adaptation, occurring
not only in sagebrush but in other vegetation types as well.
Sagebrush communities provide habitat for approximately
100 bird species and 70 mammal species (Braun and others
1976). Several species of lizards and snakes also inhabit
these sagebrush areas (Fautin 1946).
Grazers, browsers, and seedeaters foraging within sagebrush- grass communities use the grasses, forbs, sagebrush,
and/or other shrub species found there. In turn, many of
189
McAdoo, Swanson, Schultz, and Brussard
these species, including ungulates, rodents, hares and rabbits, small birds, reptiles, and insects are important as prey
for predatory species living in or near sagebrush-grass
communities.
Habitat Requirements of Sagebrush
Obligates ______________________
Because of a West-wide decline in sage grouse populations
and habitat, the sage grouse has been petitioned for listing
as threatened or endangered, and much political attention
has therefore been focused on this species. Sage grouse
historically inhabited much of the western United States,
including portions of 16 States and along the southern
border of three western Canadian Provinces. This distribution closely parallels the range of sagebrush communities.
The current core of sage grouse populations includes areas
of Nevada, Oregon, Idaho, Colorado, Wyoming, and Montana, with remnant populations in other States. These birds
require sagebrush for food and/or cover during each stage of
their life cycle (Connelly and others 2000; Klebenow 2001).
Although sage grouse depend on sagebrush vegetation for
survival, they thrive best in areas with a mixture of sagebrush species, varying in age and cover classes. Optimal
habitat includes a heterogeneous combination of diverse
sagebrush communities, in other words, sagebrush stands
with varying shrub heights and canopy cover and a diverse
understory of perennial grasses and forbs. The proportion of
sagebrush, perennial grasses, and forbs in an area varies
with the species or subspecies of sagebrush, the ecological
potential of the site, and condition of the habitat (Klebenow
2001). During the course of a year, sagebrush is quantitatively the most important component in the diet of sage
grouse, comprising 60 to 80 percent of all food consumed.
However, during spring and summer these birds (especially
the juveniles) shift from a sagebrush-dominated diet to one
of forbs and insects (Klebenow and Gray 1968).
Habitat requirements differ among the other sagebrush
obligates. Sage sparrows, Brewer’s sparrows, and sage
thrashers all require sagebrush for nesting, with nests
typically located in the sagebrush canopy. Sage thrashers
usually nest in tall dense clumps of sagebrush within areas
having some bare ground for foraging. Sage sparrows prefer
large continuous stands of sagebrush, and Brewer’s sparrows are associated closely with sagebrush habitats having
abundant scattered shrubs and short grass (Page and Ritter
1999).
Pygmy rabbits live in areas with clumps of tall sagebrush
in friable soils, whereas pronghorn antelope prefer lower
growing sagebrush, presumably because of their keen eyesight adapted for detecting danger at long distances. Like
sage grouse, both pygmy rabbits and pronghorns typically
eat sagebrush almost exclusively during winter (Page and
Ritter 1999). However, pronghorns depend primarily on
forbs during much of the year. Sagebrush voles feed on green
herbaceous plants in summer. They use the shredded bark
of sagebrush to line their burrows and eat the bark and twigs
of sagebrush during winter. Sagebrush lizards prey on
190
Vegetation Management for Sagebrush-Associated Wildlife Species
insects found in sagebrush habitat, often climbing the shrubs
in search of their prey.
Habitat Requirements of Other
Sagebrush-Associated Species ____
Habitat requirements differ widely among other bird
species associated with sagebrush grass communities. Some
species, such as loggerhead shrikes (Lanius ludovicianus),
nest primarily in the canopy of sagebrush and other shrubs.
Others are primarily open ground and/or grass nesting
species, requiring varying amounts of herbaceous cover.
Such species include horned larks (Eremophila alpestris),
vesper sparrows (Pooecetes gramineus), and western meadowlarks (Sturnella neglecta). Other species, like lark sparrows (Chondestes grammacus), are typically most abundant
in areas with a diverse mixture of sagebrush and bunchgrass
(McAdoo and Klebenow 1989). Horned larks and burrowing
owls (Athene cunicularia) are adapted to more open areas,
and both species often increase after wildfire or other disturbances that reduce dense sagebrush canopies.
In addition to pronghorn antelope, other ungulate big
game species are dependent on sagebrush-grass communities to some extent. Mule deer (Odocoileus hemionus) are
closely associated with sagebrush-grass communities in
much of their range. Being primarily browers, any successional vegetation changes favoring shrubs may benefit mule
deer populations. Many forbs and shrubs associated with
sagebrush communities are important in mule deer diets,
with grasses used primarily in spring. Forb use is highest in
summer, and on many mule deer ranges, big sagebrush is
the staple component in winter and early spring (Kufeld and
others 1973). Elk (Cervus elaphus) generally depend on
grasses for forage throughout much of their range, but they
also eat shrubs, including big sagebrush, especially during
fall and winter (Kufeld 1973). In parts of the Great Basin, elk
use sagebrush-dominated habitats in other seasons as well.
Bighorn sheep (Ovis canadensis spp.) also use sagebrushgrass communities in some areas, especially as winter range.
Although grasses are typically the major component in the
bighorn sheep diet, shrubs are important, and big sagebrush
is a preferred shrub (McQuivey 1978).
Five species of hares and rabbits may occur in sagebrushgrass communities. The most common species in most areas
is the black-tailed jackrabbit (Lepus californicus), an opportunistic feeder that selects for succulence. Blacktailed jackrabbits eat primarily grasses and forbs until winter, when
they feed on shrubs, including the leaves and bark of big
sagebrush. During population highs, this species can cause
considerable damage to rangeland vegetation and cultivated crops (McAdoo and others 1987). Within sagebrushgrass habitats, blacktailed jackrabbits are typically associated with increasing shrub cover, whereas whitetailed
jackrabbits (L. townsendii) are associated with increasing
grass cover (Verts and Carraway 1998). Pygmy rabbits have
already been identified (above) as sagebrush obligates. Two
other rabbit species, desert cottontails (Sylvilagus audubonii)
and mountain cottontails (S. nuttallii), are also found in
some sagebrush-grass habitats.
USDA Forest Service Proceedings RMRS-P-31. 2004
Vegetation Management for Sagebrush-Associated Wildlife Species
Many rodent species (at least 28) inhabit sagebrush-grass
communities, with the deer mouse (Peromyscus maniculatus)
being typically most common. Unlike the sagebrush vole
that was mentioned above as a sagebrush obligate, deer mice
occur in a wide variety of vegetation types. Great Basin
pocket mice (Perognathus parvus) are restricted primarily to
sagebrush habitats in some areas (McAdoo and Klebenow
1979). Most rodent species are herbivores or granivores, but
differ in specific habitat affinities. Rodents in general have
a reputation for negative impacts on rangelands, but some
species, such as kangaroo rats (Dipodomys sp.), can be quite
beneficial in terms of seed dispersal and germination (McAdoo
and others 1983).
At least nine bat species may be found within sagebrush
habitats, but are more closely associated with caves, rock
crevices, and water sources. The Merriam’s shrew, an insecteater, is sometimes a relatively common small mammal
species in sagebrush basins (Ports and McAdoo 1986). Western rattlesnakes, gopher snakes, leopard lizards, horned
lizards, and other reptiles also make their homes in sagebrush habitat, but amphibians are found only near water
sources that may be surrounded by sagebrush or other
upland habitat.
Although several mammalian predators use sagebrush
habitats in search of prey, none are exclusively linked to
sagebrush. The most common predator species using sagebrush habitats include coyotes, badgers, long-tailed and
short-tailed weasels, bobcats, and mountain lions.
Vegetation Management
Implications ____________________
Because habitat requirements for the many wildlife species in sagebrush-grass communities differ by species and
even vary by season of use for many species, the spatial and
temporal variability of sagebrush habitats becomes important in vegetation management. Before European settlement, “spotty and occasional wildfire probably created a
patchwork of young and old sagebrush stands across the
landscape, interspersed with grassland openings, wet meadows, and other shrub communities” (Paige and Ritter 1999).
In drier regions, such as lower elevations in the Great Basin
(that is, the sagebrush semidesert), fire probably had less
influence. According to Miller and Eddleman (2001):
The Wyoming big sagebrush and low sagebrush cover
types, with less frequent disturbance events but slower recovery rates, and the mountain big sagebrush cover type, with
more frequent disturbance but faster recovery rates, created a
mosaic of multiple seral stages across the landscape. In addition, fire patterns were patchy, leaving unburned islands,
particularly in Wyoming big sagebrush cover types because of
limited and discontinuous fuels. Plant composition ranged
from dominant stands of sagebrush to grasslands.
The authors went on to say that much of the sagebrush
steppe ecosystem during presettlement times was probably
composed of open shrub stands with a substantial component of long-lived perennial grasses and forbs.
The wildlife sightings by early explorers were a function of
landscape ecology, the explorers’ season of travel, and the
time interval since the last fire. Based on anecdotal accounts,
USDA Forest Service Proceedings RMRS-P-31. 2004
McAdoo, Swanson, Schultz, and Brussard
species like sage grouse seemed to be locally abundant, but
regionally rare. According to Miller and Eddleman (2001),
the range occupied by sage grouse is spatially diverse and
temporally dynamic. By inference, since sage grouse distribution closely parallels the range of sagebrush communities
in North America, the same principle holds for other sagebrush-associated vertebrate wildlife species. Inherent site
potential, combined with such variables as the interspersion
of varying shrub height and canopy cover, as well as herbaceous species composition, cover, and diversity, influence
the regional diversity of wildlife species, the landscape level
distribution of these species, and the local abundance of each
species. Nothing, however, remains constant over time.
We can draw some inferences from the effects of sagebrush-grass community alteration on neotropical migrants
(songbirds). Research conducted in northern and central
Nevada (McAdoo and others 1989) showed that sagebrush
removal from large acreages had initially negative impacts
on shrub-nesting birds, especially sagebrush obligates such
as sage thrashers, sage sparrows, and Brewer’s sparrows. In
those areas where crested wheatgrass (Agropyron desertorum)
was planted after shrub removal, a corresponding increase
was observed in ground and grass-nesting species like horned
larks, western meadowlarks, and lark sparrows. However,
as successional establishment of sagebrush occurred in
these areas over time, shrub-nesting bird species returned
and grass-nesting species remained. Bird species diversity
increased as the complexity of the plant community increased (McAdoo and others1989). What are the implications of these bird population responses for wildlife species
in general as related to sagebrush habitat management?
There exists an enormous challenge throughout much of
the Intermountain West to revegetate large expanses of
cheatgrass (Bromus tectorum) monocultures with native
sagebrush-grass-forb communities that will support diverse
wildlife communities. However, another habitat condition is
perhaps being overshadowed by the cheatgrass problem.
Namely, much of the Intermountain West contains large
expanses of sagebrush habitat where shrub cover is so
dominant that herbaceous cover is almost absent, with only
sparse populations of remnant native grasses and forbs. To
improve the site productivity of these areas for seasonal use
by high profile species like sage grouse, proposals have been
made to manage portions of these areas for reduced mature
sagebrush cover, regeneration of young sagebrush, and
increased native herbaceous cover. According to Klebenow
(1969), reducing/thinning sagebrush cover in some areas
can restore the balance of forbs and grasses, thereby enhancing sage grouse habitat. Connelly and others (2000) maintain that treatments such as prescribed fire, grazing, herbicides, and mechanical treatments may be used to restore
sagebrush habitats, but caution that improper use of these
tools can also result in the degradation or loss of such
habitats.
Location, and especially size of treatments, must be carefully chosen, since wildlife species respond variously to scale
of vegetation management treatments (Paige and Ritter
1999). For example, bird species in general are affected at
the population level by vegetation management treatments
covering thousands of acres, at the home range level by
management of vegetation “stands” from 1 to thousands of
191
McAdoo, Swanson, Schultz, and Brussard
acres in size, and at the individual/pair level by treatments
on areas from less than 1 to hundreds of acres (Paige and
Ritter 1999). For sage grouse specifically, Klebenow (2001)
does not recommend sagebrush eradication over large areas,
but suggests thinning sagebrush to about 15 percent cover to
enhance forb and grass production. He also suggests that
small burns in mountain big sagebrush can create a mosaic
pattern to enhance forbs and increase sagebrush height
diversity.
In lower elevation sites (dominated by Wyoming big sagebrush) where recovery of herbaceous vegetation is slower,
fire should be used cautiously to reduce the threat of
cheatgrass invasion. Some of these areas may require seeding with adapted perennial species to compete with
cheatgrass, followed after establishment by interseeding of
native shrubs, forbs, and grasses (Klebenow 2001). Connelly
and others (2001) similarly recommend the seeding of “functional equivalents” (non-native plant species) where native
forbs and grasses are unavailable. They also caution that
prescribed fire (and fire surrogate treatments like herbicides) be used cautiously in Wyoming big sagebrush habitats, realizing that 30 years is the approximate recovery
period for Wyoming big sagebrush stands. For populations of
diverse bird species, Paige and Ritter (1999) recommend
that prescribed burns to enhance vegetation diversity be
kept relatively small (with patchy distribution), reseeded
where necessary, and protected from livestock grazing until
seeded species become established.
We propose that if such management strategies were
properly implemented, a continuum of herbaceous, herbaceous-shrub, shrub-herbaceous, and shrub-dominated habitats could be created. We think that most wildlife species on
a landscape scale would be largely benefited. Because of the
diverse habitat requirements of various wildlife species,
habitat for all sagebrush-associated wildlife species would
be present in varying amounts on a landscape scale. In other
words, creating landscape heterogeneity with multiple-aged
stands of sagebrush and varying degrees of herbaceous and
shrub cover would provide both the vertical and horizontal
vegetation components of vegetation diversity required by
diverse wildlife species. The value of each landscape parcel
for various wildlife species would change over time with the
dynamics of the initial natural or prescribed disturbance,
environmental variability of secondary plant succession,
and post-disturbance management.
State and transition models, imbedded into ecological site
descriptions, offer the best tool for analyzing vegetation
management hazards and opportunities, and determining
management options and priorities (Bestelmeyer and others
2003; Laycock 1991, 1995; West 1999). Simply put, state and
transition models reflect the idea that rangeland vegetation
exhibits several “states” (recognizable complexes of soil and
vegetation structure that are resistant to change and resilient to impacts). These models also present the idea that the
various states have ecological phases through which the
states progress over time, by pathways consisting of changes
in plant species composition or community structure. On the
other hand, almost irreversible “thresholds” or boundaries
between states can be crossed in “transitions.” These transitions are reversible through reasonable management actions
192
Vegetation Management for Sagebrush-Associated Wildlife Species
until an undesired threshold is crossed. Once such a threshold is crossed, change back toward a more desired state is
irreversible unless extensive time, effort, and money are
available to effect a change. For example, after a sagebrushgrass community has crossed the ecological threshold to
dominance by cheatgrass or noxious weeds, the transition is
virtually irreversible. But carefully built and easily understood state and transition models can help identify such
ecological transitions at earlier stages, allowing appropriate
changes in management actions to prevent the crossing of
ecologically damaging thresholds.
The highest priorities for habitat treatments should be
driven by the risk of crossing an ecological threshold (such as
weed invasion) and the opportunity to apply an effective
management treatment. Only adaptive management strategies (Macnab 1983) that follow up active vegetation management with monitoring and adjustment of strategies, if
necessary, will ensure the perpetuation of a diverse and
productive landscape. Success in establishing a heterogeneous mosaic of native plant communities across any rangeland landscape also complements sustainable rangeland
management for multiple uses in addition to wildlife.
References _____________________
Bestelmeyer, B. T.; Brown, J. R.; Havstad, K. M.; Alexander, R.;
Chavez, G.; Herrick, J. E. 2003. Development and use of state-and
transition models for rangelands. Journal of Range Management.
56: 114–126.
Braun, C. E.; Baker, M. F.; Eng, R. L.; Gashwiler, J. S; Schroeder,
M. H. 1976. Conservation committee report on effects of alteration of sagebrush communities on the associated avifauna.
Wilson Bulletin. 88: 165–171.
Connelly, J. W.; Schroeder, M. A.; Sands, A. R.; Braun, C. E. 2000.
Guidelines to manage sage grouse populations and their habitats. Wildlife Society Bulletin. 28: 967–985.
Fautin, R. W. 1946. Biotic communities of the northern desert shrub
biome in western Utah. Ecological Monographs. 16(3): 251–310.
Klebenow, D. A. 2001. Enhancing sage grouse habitat…a Nevada
landowner’s guide. Northwest Sage Grouse Working Group
Publication.
Klebenow, D. A. 1969. Sage grouse nesting and brood habitat in
Idaho. Journal of Wildlife Management. 33: 649–662.
Klebenow, D. A.; Gray, G. M. 1968. Food habits of juvenile sage
grouse. Journal of Range Management. 21: 80–83.
Kufeld, R. C. 1973. Foods eaten by Rocky Mountain Elk. Journal of
Range Management. 26: 106–113.
Kufeld, R. C.; Wallmo, O. C.; Feddema, C. 1973. Foods of the Rocky
Mountain mule deer. Res. Pap. RM-111. Fort Collins, CO: U.S.
Department of Agriculture, Forest Service, Rocky Mountain
Forest and Range Experiment Station. 31 p.
Laycock, W. A. 1991. Stable states and thresholds of range condition
on North American rangelands—a viewpoint. Journal of Range
Management. 44: 427–433.
Laycock, W. A. 1995. New perspectives on ecological condition of
rangelands: can state and transition or other models better define
condition and diversity? In: Montes, L.; Olivia, G. E., eds. Proceedings—international workshop on plant genetic resources,
desertification, and sustainability. Santa Cruz, Argentina: Centro
Regional Patagonia Sur, INTA-EEA, Rio Gallegos: 140–164.
Macnab, J. 1983. Wildlife management as scientific experimentation. Wildlife Society Bulletin. 11: 397–401.
McAdoo, J. K.; Evans, C. C.; Roundy, B. A.; Young, J. A.; Evans, R. A.
1983. Influence of heteromyid rodents on Oryzopsis hymenoides
germination. Journal of Range Management. 36: 61–64.
McAdoo, J. K.; Klebenow, D. A. 1979. Native faunal relationships in
sagebrush ecosystems. In: The sagebrush ecosystem symposium.
Logan: Utah State University: 50–61.
USDA Forest Service Proceedings RMRS-P-31. 2004
Vegetation Management for Sagebrush-Associated Wildlife Species
McAdoo, J. K.; Longland, W. S.; Cluff, G. J.; Klebenow, D. A. 1987.
Use of new rangeland seedings by black-tailed jackrabbits. Journal of Range Management. 40: 520–524.
McAdoo, J. K.; Longland, W. S.; Evans, R. A. 1989. Nongame bird
community responses to sagebrush invasion of crested wheatgrass seedings. Journal of Wildlife Management. 53: 494–502.
McQuivey, R. P. 1978. The desert bighorn sheep of Nevada. Biol.
Bull. No. 6. Nevada Department of Fish and Game. 81 p.
Miller, R. F.; Eddleman, L. L. 2001. Spatial and temporal changes
of sage grouse habitat in the sagebrush biome. Tech Bull. 151.
Corvallis: Oregon State University Agricultural Experiment Station. 35 p.
Paige, C.; Ritter, S. A. 1999. Birds in a sagebrush sea: managing
sagebrush habitats for bird communities. Boise, ID: Partners in
Western Flight Working Group.
Ports, M. A.; McAdoo, J. K. 1986. Sorex merriami (Insectivora:
Soricidae) in eastern Nevada. Southwest Naturalist. 31: 415–416.
Tisdale, E. W.; Hironaka, M. 1981. The sagebrush-grass region: a
review of the ecological literature. Bull. 33. Moscow: University
USDA Forest Service Proceedings RMRS-P-31. 2004
McAdoo, Swanson, Schultz, and Brussard
of Idaho, College of Forestry, Wildlife, and Range Sciences,
Forest, Wildlife and Range Experiment Station.
Vale, T. R. 1975. Presettlement vegetation in the sagebrush-grass
area of the Intermountain West. Journal of Range Management.
28: 32–36.
Verts, B. J.; Carraway, L. N. 1998. Land mammals of Oregon.
Berkeley: University of California Press.
West, N. E. 1983a. Great Basin-Colorado Plateau sagebrush semidesert. In: West, N. E., ed. Ecosystems of the World 5. Temperate
deserts and semi-deserts. New York: Elsevier. Chapter 12 .
West, N. E. 1983b. Western Intermountain sagebrush steppe. In:
West, N. E., ed. Ecosystems of the World 5. Temperate deserts
and semi-deserts. New York: Elsevier. Chapter 13.
West, N. E. 1999. Synecology and disturbance regimes of sagebrush
steppe ecosystems. In: Entwistle, P. G.; DeBolt, A. M.; Kaltenecker,
J. H.; Steenhof, K., comps. Proceedings: sagebrush steppe ecosystems symposium. Boise, ID: Boise State University: 15–26.
193
Browsing Effects on Wyoming
Big Sagebrush Plants and
Communities
Carl L. Wambolt
Trista Hoffman
Abstract: The effect of likely yearlong browsing by several wild
ungulate species on individual Wyoming big sagebrush (Artemisia
tridentata ssp. wyomingensis) plants and communities was studied.
The investigation was conducted near Gardiner, MT, in the ungulate-rich boundary line area of the Northern Yellowstone Winter
Range. Plant level responses were measured in this study and
related to reported community responses. Individual sagebrush
plants were significantly different (P < 0.05) in and out of an
exclosure 35 years after it was constructed (1957) in regard to
production, seedhead number, and leaf dry weight. Certain morphological characters of the same plants were not impacted by browsing. These results relate well to previously detailed plant community responses on this important ungulate range. The Wyoming big
sagebrush has been greatly reduced in the study area by browsing.
Introduction ____________________
Big sagebrush (Artemisia tridentata) taxa are particularly important for ungulates as forage and often security
and/or thermal cover in portions of the Northern Yellowstone
Winter Range (NYWR) that are relatively free of snow.
Under these conditions, the plants remain accessible for
foraging throughout winter (Wambolt 1998). Twentiethcentury naturalists (Cahalane 1943; Kittams 1950; Rush
1932; Wright and Thompson 1935) have commented on the
conspicuous use of sagebrush for forage and cover by ungulates on the NYWR and expressed concern over what they
considered excessive use of sagebrush in the winter diets of
ungulates in Yellowstone National Park (YNP).
The effects of Yellowstone’s large populations of ungulates
on sagebrush taxa on the NYWR have been debated for more
than 70 years (Rush 1932; Wright and Thompson 1935). The
National Park Service (NPS) was concerned enough about
sagebrush and other browse on the NYWR that they constructed ten 2-ha exclosures on the NYWR in 1957 and 1962
to investigate the relationships between ungulate foraging
and plant communities.
We used a NPS exclosure erected in 1957 to test the
hypothesis that Wyoming big sagebrush plants protected
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Carl L. Wambolt is a Professor of Range Science and Trista Hoffman is a
Research Assistant, Montana State University, P.O. Box 172900, Bozeman,
MT 59717, U.S.A. FAX: (406) 994-5589, e-mail: cwambolt@montana.edu
194
from browsing for 35 years would exhibit growth characteristics similar to browsed plants. These results were then
related to plant community responses on the NYWR.
Study Area and Methods _________
The natural winter range provided by the Gardiner Basin
is created by the orographic effects on precipitation of mountain peaks up to 3,353 m in elevation. Ideal winter range on
extensive south- or west-facing aspects in the Gardiner
Basin have been influenced by glacial scouring, morainal
deposition, and outwash sediments. In the portion of the
NYWR where this study was conducted, the sagebrush
habitat is the Wyoming big sagebrush (Artemisia tridentata
ssp. wyomingensis)-bluebunch wheatgrass (Agropyron spicatum)
type.
Individual plant characteristics (table 1) were sampled
from 20 plants randomly selected and then paired by size in
and out of the 1957 exclosure. Seedheads were counted. Ten
leaders were clipped per plant. Leaves were counted and
dried at 65 ∞C, and dry weight was determined for each
leader and its leaves. The Kolmogorov-Smimov test compared differences in distributions, and the Wilcoxon test
compared distribution medians.
Sampling of community characteristics were made in and
out of the above described exclosure (1957 construction) and
a second similar exclosure constructed in 1962 approximately 300 yards from the first exclosure. This sampling
included measurements of individual sagebrush plants for
seedhead number and length, average cover, major and
minor axes, crown depth, and overall height. These measurements estimated sagebrush productivity from models
previously developed from Wyoming big sagebrush plants in
Table 1—Average differences between browsed and unbrowsed
Wyoming big sagebrush plants.
Characteristic
Production (g/plant)
Seedheads per plant
Leader length (mm)
Leader dry weight (g)
Leaves per leader
Leaf area (mm2)
Leaf dry weight (g)
a
Browseda
10.0a
0.08a
22.9a
.02a
43.0a
354.0a
0.06a
Unbrowseda
44.7b
60.3b
22.3a
.02a
44.0a
354.0a
0.08b
Values followed by different letters are significant (P < 0.05).
USDA Forest Service Proceedings RMRS-P-31. 2004
Browsing Effects on Wyoming Big Sagebrush Plants and Communities
Wambolt and Hoffman
the Gardiner Basin (Wambolt and others 1994). Additional
production information is reported in Wambolt and Sherwood
(1999). Shrub density and line intercept (canopy cover) were
determined for both young and old plants. Because the 2-ha
exclosures contained considerable environmental variation,
the sagebrush habitat within each exclosure was stratified
(Hurlbert 1984) by separating topographic, soil, and microclimatic variation into eight paired sites. With random
sampling using paired sites (in and out of the exclosures), it
is unlikely that comparable distributions of topo-edaphic
positions would have been obtained regardless of sample
size (Coughenour 1991). Sampling areas were paired in and
out of each exclosure within four slope-aspect combinations
(table 2). These areas were compared with students’ t-tests.
Discussion _____________________
Because of heavy browsing, plants outside the exclosure
had no terminal leader growth. However, the plants inside
the exclosure were dominated by terminal growth, and axial
long shoots were rare. Thus, for further investigation, it was
necessary to compare the terminal leaders of protected
plants inside the exclosure to axial long shoots on browsed
plants.
Unbrowsed plants had consistently higher production
than browsed plants (table 1). The average production per
plant was 10 g with browsing and 45 g with protection. No
measurements of dead crown were taken, but plants under
protection appeared vigorous, whereas plants outside the
exclosure had large amounts of dead crown.
The greatest difference between browsed and unbrowsed
plants was in seedhead production. Seedheads overaged
0.08 per browsed plant and 60.3 per unbrowsed plant.
Related studies confirm that stress, such as herbivory, may
delay or prevent flowering for several years (Bazzaz 1987;
Maschinski and Whitham 1989; McConnell and Smith 1977).
The lower overall production of browsed sagebrush plants
indicates that foliage loss results in reduced reproductive
potential. Bilborough and Richards (1991) found that buds
for flowering stems on mountain big sagebrush were located
on short shoots at the distal end of the terminal leader.
Because almost all terminal leaders were removed on browsed
plants, flowering stems would have to be initiated from
elsewhere. The loss in seedhead production from browsing
on the NYWR has undoubtedly resulted in declines in
reproduction for sagebrush because the taxon lacks any
asexual means of reproduction.
A companion study found that the addition of average
seedhead weight improved the capability of models to
predict production of winter forage from the three NYWR
big sagebrush subspecies (Wambolt and others 1994). However, these improved models could only be used for two of
the taxa when they exhibited light use (browse form class).
Heavily used plants produced few inflorescences; therefore, the addition of average seedhead weight to the model
was most useful for predicting forage production from lowuse plants (Wambolt and others 1994). The segregation of
browse form classes and inclusion of average seedhead
weight in the models acknowledge the impact browsing has
had on the annual production and reproduction of NYWR
sagebrush and relate well to our findings of individual
plant characteristics.
Thirty-five percent of mountain big sagebrush (A. t.
vaseyana) plants were killed by heavy browsing between
1982 and 1992 (Wambolt 1996). Many surviving plants
developed a heavy-use browse form class with a high percentage of dead crown. The dead crown in the three big
Table 2—Percent canopy cover of Wyoming big sagebrush and number of big sagebrush plants (with a
minimum canopy of 15 cm) per 60 m2 at eight environmentally paired sites, either browsed or
protected. The paired sites are associated with the exclosures established either in 1957 or
1962 (Wambolt and Sherwood 1999).
Site
Slopea
Aspect
Protected
Browsed
Probability > t
Gardiner-57A
Gardiner-57B
Gardiner-57C
Gardiner-57D
Gardiner-62A
Gardiner-62B
Gardiner-62C
Gardiner-62D
Flat
Steep
Gentle
Gentle
Moderate
Flat
Very steep
Very steep
Flat
E
SW
NWW
NEE
Flat
NEE
SE
- - - - - Canopy cover (percent)- - - - 3.9
0.0
0.0001
3.6
.1
.0004
4.5
1.1
.0001
1.4
.4
.0073
21.8
.4
.0001
17.6
4.3
.0000
2.4
.2
.0012
6.8
.0
.0001
Gardiner-57A
Gardiner-57B
Gardiner-57C
Gardiner-57D
Gardiner-62A
Gardiner-62B
Gardiner-62C
Gardiner-62D
Flat
Steep
Gentle
Gentle
Moderate
Flat
Very steep
Very steep
Flat
E
SW
NWW
NEE
Flat
NEE
SE
- - - - - - Density (per 60 m2) - - - - - 9.0
.5
.0002
8.1
.5
.0002
15.1
5.0
.0005
7.7
1.0
.0001
36.0
1.2
.0001
39.2
.2
.0001
2.2
.0
.0090
6.9
.7
.0001
a
Slope classes are: flat < 3 percent, gentle = 4 to 15 percent, moderate = 16 to 29 percent, steep = 30 to 44 percent,
very steep ≥ 45 percent.
USDA Forest Service Proceedings RMRS-P-31. 2004
195
Wambolt and Hoffman
sagebrush subspecies increased in proportion to the overall
amount of browsing received by each taxon. The percentage
of dead crown in live plants for mountain big sagebrush,
Wyoming big sagebrush, and basin big sagebrush (A. t. ssp.
tridentata) was 58.7, 45.4, and 30.1, respectively (Wambolt
1996).
A significant difference was found between the development of protected and browsed big sagebrush communities
(table 2). Wambolt and Sherwood (1999), who studied mountain big sagebrush in addition to Wyoming big sagebrush,
stated:
Since the period of exclosure construction in 1957 and 1962,
there has been a significant difference in the development of
protected and browsed big sagebrush communities. Average
big sagebrush canopy cover on protected sites was 202 percent
greater (P < 0.0027) than on browsed sites over the nineteenpaired sites. The average big sagebrush cover for all 19 sites
was 19.7 percent inside and 6.5 percent outside the exclosures. This relationship was universal on sites with Wyoming
big sagebrush or mountain big sagebrush, flat to very steep
topographies, and all aspects and precipitation levels.
However, the differences between in and out of the exclosures at the Wyoming big sagebrush sites discussed in this
paper were much greater than at the mountain big sagebrush locations. The Wyoming big sagebrush sites under
protection averaged almost 10 times more sagebrush cover
than where browsing had continued following exclosure
construction. Ungulate browsing also affected numbers of
big sagebrush plants. Across the NYWR, big sagebrush
plants were twice as numerous with protection as with
browsing.
The production differential for Wyoming big sagebrush
was great, but because browsed plants were so reduced in
size, their growth parameters were not suitable for production models (Wambolt and others 1994). The response of
the sprouting shrubs, green rabbitbush (Chrysothamnus
visidiflorus), rubber rabbitbrush (C. nauseosus), and gray
horsebrush (Tetradymia canescens), as measured by canopy
cover and density, was similar to that of big sagebrush across
the NYWR.
Singer and Renkin (1995) and Wambolt and Sherwood
(1999) found a large impact from browsing on the Wyoming
big sagebrush in the Gardiner Basin (fig. 1). Wambolt and
Sherwood (1999) considered the difference in impact between Wyoming big sagebrush and mountain big sagebrush
with this statement: “Pronghorn and mule deer often forage
heavily on big sagebrush taxa (Welch and McArthur 1979).
Mule deer diets averaged 52 percent big sagebrush over a 10
year period (Wambolt 1996), only a couple of kilometers
away from the eight Wyoming big sagebrush paired sites. A
high degree of utilization is reflected in the great impact on
Wyoming big sagebrush populations at these eight sites
where elk may also be present with pronghorn and mule deer
(Singer and Renkin 1995).”
Wambolt (1996) concluded that any of the four NYWR
sagebrush taxa would be heavily browsed if severe winter
conditions precluded ungulates from exercising their preferences. However, mountain big sagebrush was clearly the
preferred taxon by mule deer and elk. The fact that Singer
and Renkin (1995) and Wambolt and Sherwood (1999)
found Wyoming big sagebrush to be more impacted than
196
Browsing Effects on Wyoming Big Sagebrush Plants and Communities
mountain big sagebrush was largely a function of snow
depth limiting pronghorn and mule deer foraging over the
larger distribution of mountain big sagebrush. The intensive competition from a large elk population also restricted
the winter range of the smaller ungulates to the Wyoming
big sagebrush habitat type.
Conclusions ____________________
The historical evidence and recent studies (Patten 1993;
Wambolt 1998; Wambolt and Sherwood 1999) indicate a
significant decline of NYWR sagebrush. This potentially
impacts ungulates that rely on sagebrush habitat for meeting their nutritional needs and other requirements. Sagebrush taxa are highly nutritious and preferred forage for
ungulates (Welch and McArthur 1979). They have been bred
and selected to improve the forage values of rangelands
(Welch and Wagstaff 1992). Sagebrush is particularly high
in protein. Welch and McArthur (1979) found the midwinter
crude protein content of 21 big sagebrush accessions averaged 12.4 percent (range = 10 to 16 percent).
Ungulates such as pronghorn are impacted as the loss of
their highest winter protein source (sagebrush) continues
(Welch and McArthur 1979). This loss will manifest itself in
a decreasing ability to meet their nutritional needs and
requirements for reproduction. The implications for other
organisms are clear. In concert with the decline of the native
vegetation (sagebrush and dependent species), it is reasonable to expect that numerous animals (Welch 1997) will be
impacted.
The following points summarize our findings:
1. Wyoming big sagebrush shrubs that were protected
from ungulate browsing produced significantly more forage
than browsed shrubs.
2. Seedhead production was severely reduced by browsing.
Heavily browsed stands may have difficulty regenerating.
3. Leaders on browsed shrubs have the same number of
leaves as those on unbrowsed shrubs, but the mass of leaves
is reduced.
4. Shrub composition of the Wyoming big sagebrush habitat type has been significantly changed by browsing on the
NYWR.
5. Big sagebrush may be overused and severely damaged
on big game winter ranges.
References _____________________
Bazzaz, F. A.; Chiariello, N. R.; Colley, P. D.; Ptielka, L. F. 1987.
Allocating resources to reproduction and defense. Bioscience. 37:
58–67.
Bilbrough, C. J.; Richards, J. H. 1991. Branch architecture of
sagebrush and bitterbrush: use of a branch complex to describe
and compare patterns of growth. Canadian Journal of Botany. 69:
1288–1295.
Cahalane, V. H. 1943. Elk management and their regulation—
Yellowstone National Park. Transactions of the North American
Wildlife Conference. 8: 95–101.
Coughenour, M. B. 1991. Biomass and nitrogen responses to grazing
of upland steppe on Yellowstone’s northern winter range. Journal
of Applied Ecology. 28:71–82.
Hurlbert, S. H. 1984, Pseudoreplication and the design of ecological
field experiments. Ecological Monographs. 54: 187–211.
USDA Forest Service Proceedings RMRS-P-31. 2004
Browsing Effects on Wyoming Big Sagebrush Plants and Communities
Kittams, W. H. 1950. Sagebrush on the lower Yellowstone range as
an indicator of wildlife stocking. Unpublished report on file at:
Yellowstone National Park.
Maschinski, J.; Whitham, T. J. 1989. The continuum of plant
responses to herbivory: the influence of plant association, nutrient availability and timing. The American Naturalist. 134: 1–19.
McConnell, B. R.; Smith, J. G. 1977. Influence of grazing on ageyield interactions in bitterbrush. Journal of Range Management.
30: 91–93.
Patten, D. T. 1993. Herbivore optimization and overcompensation:
does native herbivory on western rangelands support these
theories? Ecological Applications. 3: 35–36.
Rush, W. M. 1932. Northern Yellowstone elk study. Helena: Montana Fish and Game Commission.
Singer, F. J.; Renkin, R. A. 1995. Effects of browsing by native
ungulates on the shrubs in big sagebrush communities in
Yellowstone National Park. Great Basin Naturalist. 55: 201–212.
Wambolt, C. L. 1996. Mule deer and elk foraging preference for 4
sagebrush taxa. Journal of Range Management. 49: 499–503.
Wambolt, C. L. 1998. Sagebrush and ungulate relationships on Yellowstone’s northern range. Wildlife Society Bulletin. 26: 429–437.
USDA Forest Service Proceedings RMRS-P-31. 2004
Wambolt and Hoffman
Wambolt, C. L.; Creamer, W. H.; Rossi, R. J. 1994. Predicting big
sagebrush winter forage by subspecies and browse form class.
Journal of Range Management. 47: 231–234.
Wambolt, C. L.; Sherwood, H. W. 1999. Sagebrush response to
ungulate browsing in Yellowstone. Journal of Range Management. 52: 363–369.
Welch, B. L. 1997. Seeded versus containerized big sagebrush
plants for seed-increase gardens. Journal of Range Management.
50: 611–614.
Welch, B. L.; McArthur E. D. 1979. Feasibility of improving big
sagebrush (Artemisia tridentata) for use on mule deer winter
ranges. In: Goodin, J. R.; Northington, D. K., eds. Arid land plant
resources. Lubbock: Texas Tech University: 451–453
Welch, B. L.; Wagstaff, F. J. 1992. “Hobble Creek” big sagebrush vs.
antelope bitterbrush as a winter forage. Journal of Range Management. 45: 140–142.
Wright, G. M.; Thompson, B. H. 1935. Fauna of the National Parks
of the United States: wildlife management in the National Parks.
Fauna Series 2. National Park Service.
197
Effects of Wildlife Utilization and
Grass Seeding Rates on Big
Sagebrush Growth and Survival
on Reclaimed Mined Lands
Kristene A. Partlow
Richard A. Olson
Gerald E. Schuman
Scott E. Belden
Abstract: Ensuring Wyoming big sagebrush (Artemisia tridentata
Nutt ssp. wyomingensis Beetle & Young) survival remains a challenge years after initial re-establishment on reclaimed mined
lands. Wildlife utilization of big sagebrush can be a major factor
influencing its survival. A wildlife-proof exclosure was erected on a
portion of an existing sagebrush establishment research site initiated by Schuman and others (1998) in 1990 at the North Antelope/
Rochelle Complex mine in northeastern Wyoming. Investigations
focused on the effects of wildlife utilization of big sagebrush growth
and survival as affected by grass seeding rates of the original study
and the newly constructed exclosure. Results indicate no significant
differences in big sagebrush density between grass seeding rates or
inside versus outside the exclosure. Significantly greater leader
growth of big sagebrush occurred inside compared to outside the
exclosure. Mean leader length of big sagebrush inside the exclosure
in April 2002, 10 months after construction, was 46.4 mm compared
to 9.4 mm outside. Wildlife browsing occurred on 100 percent of the
big sagebrush plants outside the exclosure in 2002. Utilization and
mortality of the sagebrush plants was significantly higher at the
lower grass seeding rates. Approximately 33 percent of the studied
sagebrush plants outside of the exclosure died during the 15-month
study period compared to 11 percent inside the exclosure. Findings
of this study indicate that wildlife browsing on these sagebrush
plants is significantly influencing their survival and growth.
Introduction ___________________
Wyoming big sagebrush (Artemisia tridentata Nutt ssp.
wyomingensis Beetle & Young), if present in premined
ecosystems, is required to be re-established according to the
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Kristene A. Partlow is a Research Assistant and Richard A. Olson is a
Professor at the University of Wyoming, Department of Renewable Resources, Laramie, WY 82071-3354, U.S.A., e-mail: rolson@uwyo.edu. Gerald
E. Schuman is a Soil Scientist at the USDA-ARS High Plains Grasslands
Research Station, Cheyenne, WY 82009, U.S.A. Scott E. Belden is a Senior
Environmental Supervisor at Powder River Coal Co., North Antelope/Rochelle Complex, Peabody Energy Co., Gillette, WY 82716, U.S.A.
198
Surface Mining Control and Reclamation Act of 1977 and the
Wyoming Environmental Quality Act of 1973 (Wyoming
Department of Environmental Quality, Land Quality Division 1996). The process of re-establishing this shrub has
been difficult for reclamation specialists. In 1990, Schuman
and others (1998) initiated a study to evaluate the effects of
various topsoil management, mulch type, and grass seeding
rate treatments on re-establishment of Wyoming big sagebrush. The direct-placed topsoil treatment produced a higher
big sagebrush seedling density than stockpiled topsoil in the
first 2 years of the study. They believed this was due to
consistently higher soil moisture in direct-placed topsoil
plots. Their study demonstrated the positive benefits of
direct-placed topsoil compared to stockpiled topsoil and
various mulch treatments on big sagebrush re-establishment at North Antelope Coal Mine south of Gillette, WY.
However, ensuring big sagebrush survival remains a
challenge years after initial re-establishment. Reclamation
specialists are exploring other potential impacts to big
sagebrush survival beyond edaphic and vegetative factors.
Impacts of wildlife browsing may be a major factor on big
sagebrush survival for some mines. Newly reclaimed coal
mined lands often provide young, highly palatable and
nutrient-rich plant communities that attract wildlife species such as mule deer (Odocoileus hemionus), pronghorn
antelope (Antilocapra americana), cottontail rabbits
(Sylvilagus audubonii baileyi), and jackrabbits (Lepus townsendii and Lepus californicus melanotis). Big sagebrush is a
major diet component of many wildlife species because it
provides a critical source of winter browse and cover (Beetle
1960). Since adjacent native rangelands usually contain
older, mature shrubs of lower palatability and nutrient
value, wildlife are attracted to reclaimed areas supporting
greater herbaceous vegetation. Cool-season grasses and
some shrub species, including big sagebrush, generally dominate reclamation seeding mixtures. The restriction of public
access and prohibited hunting on mine property provides an
environment that encourages habitual wildlife utilization of
these reclaimed areas.
To investigate the influence of wildlife utilization on big
sagebrush growth and survival, a wildlife-proof exclosure
was constructed on half of the original North Antelope study
site to provide comparative data on browsed versus unbrowsed
big sagebrush, therefore, providing data on browsing impacts.
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Wildlife Utilization and Grass Seeding Rates on Bg Sagebrush …
Browsing on vegetation can be positive, such as compensatory growth due to moderate use (McNaughton 1983), or
negative, such as increased mortality from heavy use
(Wambolt 1996).
Reclamation specialists have learned to successfully reestablish big sagebrush on reclaimed lands; however, developing successful management practices to ensure survival of
these new seedlings must be addressed. Factors such as low
seedling vigor, their inability to compete with herbaceous
species, poor seed quality, and altered edaphic conditions
can impact initial establishment and long-term survival of
sagebrush (Cockrell and others 1995).When these factors
are combined with intense wildlife browsing, sustaining big
sagebrush re-establishment is challenging. Quantitative
information on big sagebrush utilization and the effects of
browsing impacts on long-term seedling survival are needed.
Specific objectives of this project were to (1) determine longterm big sagebrush survival using data from the original
study; (2) establish a new baseline data record of big sagebrush survival without wildlife influence using inside versus outside exclosure comparisons; (3) assess sagebrush
density (plants m–2) within various grass seeding rates,
inside and outside the exclosure; (4) determine percent of big
sagebrush plants browsed by grass seeding rates, inside and
outside the exclosure; (5) examine seasonal (spring/summer,
fall/winter) utilization rates of big sagebrush leader growth
among grass seeding rates, inside/outside the exclosure; and
(6) recommend potential management practices to enhance
long-term big sagebrush survival on reclaimed mine lands.
Study Area ____________________
North Antelope Coal Mine is located in the Powder River
Basin in northeast Wyoming, approximately 100 km south
of Gillette, WY. Elevation ranges from 1,220 to 1,520 m.
Climate is characterized as semiarid, temperate, and continental. Average annual temperature is 7 ∞C with January
the coldest month (–6 ∞C) and July the warmest month (22
∞C). Annual precipitation is 339 mm (1978 to 2000 average),
with the greatest precipitation occurring in April, May, and
June (Schuman and Belden 2002). The frost-free growing
season averages 133 days (Glassey and others 1955).
The project area is approximately 1.2 ha in size. Native
soils are broadly classified as Haplargids, Natrargids, and
Torrorthents. Loamy and clayey soil textures comprise much
of the basin (Young and Singleton 1977). Fresh direct-placed
topsoil used at the research site included a complex of
Shingle (loamy, mixed, calcareous, mesic, shallow, Ustic
Torrorthents) and Samsil (clayey, montmorillinitic, calcareous, mesic, shallow, Ustic Torrorthents) series (Schuman
and others 1998).
Topography is characterized as plains and low-lying irregular hills. Vegetation consists of low-growing shrubs,
forbs, and short to midgrasses, primarily cool-season perennials. Prior to mining, the native vegetation primarily consisted of western wheatgrass (Pascopyrum smithii (Rybd.)
A. Love), needleandthread grass (Stipa comata Trin. and
Rupr.), prairie junegrass (Koeleria macrantha L. Pers.), big
sagebrush, sandberg bluegrass (Poa secunda Presl), sixweeks-grass [Vulpia octoflora (Walt.) Rydb.], and cheatgrass
USDA Forest Service Proceedings RMRS-P-31. 2004
Partlow, Olson, Schuman, and Belden
(Bromus tectorum L.) (Western Water Consultants and
Bureau of Land Management 1998). Species seeded on the
reclaimed area were a mixture of native cool-season perennial grasses (Schuman and others 1998), including “Rosana”
western wheatgrass [Pascopyrum smithii (Rybd.) A. Love],
“San Luis” slender wheatgrass [Elymus trachycaulus (Link)
Gould ex Skinner], and “Critana” thickspike wheatgrass
[Elymus lanceolatus (Scribner & J.G. Smith) Gold].
The predominant land use in the area is domestic livestock grazing and habitat for big game, predators, small
mammals, upland game birds, and nongame birds.
Methods and Materials __________
Experimental Design
The original big sagebrush re-establishment study design
(Schuman and others 1998) initiated in August 1990 was
used for this project and included the following treatments:
topsoil management (fresh stripped/direct-placed and 5-yearold stockpiled topsoil), mulch type (stubble mulch, surfaceapplied straw mulch, stubble and surface-applied straw
mulch, and no mulch), and grass seeding rate (no perennial
grass seeded, 16 kg PLS [pure live seed] ha–1, and 32 kg PLS
ha–1). All treatments were randomly assigned to a randomized block, split-split plot design with three replications
(fig. 1). Topsoil treatment plots were 15 by 60 m with mulch
subplots measuring 15 by 15 m and grass seeding rate subsubplots measuring 15 by 5 m within each topsoil treatment main plot. Each of the four mulch types occurred within
each of the three replications of fresh and stored topsoil
treatments. The three grass seeding rates were randomly
established within each of the four mulch treatments. Nine
quadrats (1 m2) were permanently staked in each of the
grass seeding rate sub-subplots in three belts of three
quadrats, lying in an east-west direction and located 1 m
from the edge of each subplot.
In the original study, the direct-placed topsoil plots supported higher big sagebrush density than stockpiled topsoil
in the first 2 years. Greater sagebrush seedling establishment in the direct-placed topsoil was due to consistently
higher soil moisture observed in that treatment (Schuman
and others 1998). Direct-placed topsoil also exhibits better
chemical, physical, and biological properties than stockpiled
topsoil (DePuit 1988; Stahl and others 1988; White and
others 1989). The stockpiled topsoil treatment was excluded
from this study because of the noted benefits of direct-placed
topsoil. Stockpiled topsoil also no longer represented the
original qualities because of the rapid inoculation by arbuscular
mycorrhizae via wind and the many positive biogeochemical
and physical changes that have occurred in the soil over 12
years.
Another addition to the original study was the construction of a wildlife-proof exclosure in June 2001 just prior to
the first data collection. Exclosure dimensions are 90 by
30 m and 3.1 m tall, which enclosed half of each replicated
direct-placed topsoil treatment plot. The same number of
mulch treatment subplots and grass seeding rate sub-subplots were located inside and outside the exclosure. The
exclosure fence was constructed of woven wire with 0.5 m
199
Partlow, Olson, Schuman, and Belden
Effects of Wildlife Utilization and Grass Seeding Rates on Big Sagebrush …
1
Stubble
3
+
2
Surface
mulch
2
2
3
1
3
Stubble
1
+
2
Surface
mulch
3
3
Surface
mulch
2
3
2
Surface
mulch
1
1
1
1
1
1
2
Stubble
mulch
2
2
Stubble
mulch
3
3
3
3
2
2
2
Control
1
STS
1
1
DTS
Surface
mulch
Control
Stubble
3
Control
3
DTS
Stubble
mulch
STS
+
Surface
mulch
DTS
STS
Grass seeding rates:
1 = 0 kg PLS ha
–1
2 = 16 kg PLS ha–1
3 = 32 kg PLS ha
–1
DTS = direct-placed topsoil
STS = stockpiled topsoil
Dark outline area = exclosure
Total Area = 90 x 60 m
Figure 1—Study plot design including topsoil, mulch, and grass seeding
rate treatments, North Antelope Coal Mine, Gillette, WY, 2001–2002.
high chicken wire extending along the ground surface to
exclude rabbits and large rodents.
Big Sagebrush Density
Two methods of determining big sagebrush density were
used. Permanent quadrats established in 1992 (Schuman
and others 1998) were sampled to determine long-term big
sagebrush survival. The number of live big sagebrush plants
was counted in each of nine permanent quadrats within each
grass seeding rate sub-subplot. Density was reported as the
mean number of plants m–2 in each grass seeding rate inside
and outside the exclosure.
Big sagebrush density was also assessed along a 2- by 12-m
belt transect in each grass seeding rate sub-subplot, which
provided another baseline record for evaluating big sagebrush survival inside and outside the exclosure. A 1-m2
200
quadrat was placed alongside each belt transect, sagebrush
plants were counted within the quadrat, then the quadrat
was flipped to the next meter of the transect until all
sagebrush plants were counted in the 24-m2 area of the belt
(12 m2 on each side of the line). Density was summarized as
the mean number of plants m–2 in each grass seeding rate
inside and outside the exclosure.
Percent Browsed Big Sagebrush Plants
Within each grass seeding rate sub-subplot, four big
sagebrush plants were selected near the outside corners of
the permanent quadrats making a total of 144 plants (72
outside and 72 inside the exclosure). Each selected plant was
marked by attaching a plastic locking zip tie at the base of
the plant. Plant locations were mapped for easy relocation
during subsequent sampling periods of the project. Each
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Wildlife Utilization and Grass Seeding Rates on Bg Sagebrush …
10
8
0
7
16
6
32
5
4
3
2
1
0
1992
Big Sagebrush Leader Utilization
Differences in big sagebrush leader lengths of individually
marked plants were evaluated to assess the degree of utilization. All leaders on marked plants were measured to the
nearest 1 mm, and summarized as mean leader length per
plant for each grass seeding rate inside and outside the
exclosure. Nondistinctive leaders, such as those lacking
woodiness or leaves extending from primary stems, were not
considered leaders. Leader measurements excluded leaves
extending from the terminal tip.
Big sagebrush leaders were measured in the spring and
fall each year. Percent utilization was only calculated for the
marked plants outside the exclosure because no browsing
was observed inside the exclosure. The difference in mean
leader length from spring to fall outside the exclosure provided percent summer utilization. Before new annual growth
appeared the following spring, leader length was reassessed
to provide the percent utilization during the late fall and
winter period.
Grass seeding rate
9
Density (plants m–2)
marked plant was inspected during sampling dates and
recorded as browsed or unbrowsed to estimate percent of
plants browsed.
Since the first sampling period (June 2001) immediately
followed exclosure construction, some browsing had occurred prior to exclosure erection. Therefore, percent
browsed plants were calculated inside and outside the
exclosure in June 2001. In fall 2001, a small hole in the
exclosure was discovered and repaired. Therefore, rabbit
presence and potential utilization of sagebrush leaders was
possible inside the exclosure until September 2001.
Partlow, Olson, Schuman, and Belden
1994
1996
1998
2000
2002
Year
Figure 2—Historic and present mean big sagebrush
density by grass seeding rates (kg PLS ha–1) from
permanent quadrat sampling, North Antelope Coal
Mine, Gillette, WY.
for grass seeding rates over time. Future studies should
evaluate long-term grass seeding rate influences on big
sagebrush density and survival inside the exclosure where
wildlife influences will not be a factor.
When examining 2001–2002 data only, mean big sagebrush density within the belt transects was highest in the 32
kg PLS ha–1 grass seeding rate inside the exclosure compared to other grass seeding rates, and significantly greater
than the big sagebrush density in the same grass seeding
rate outside of the exclosure (fig. 3). Big sagebrush density
also exhibited a significant location by sample date interaction
(fig. 4). Big sagebrush density was significantly higher inside
Data Analysis
Results and Discussion _________
Big Sagebrush Density
Big sagebrush density (plants m–2), measured in the
permanent quadrats, increased in 1993 and 1994 following
the 1992 seeding, but declined during subsequent years
across all grass seeding rates (fig. 2) and mulch treatments
(Schuman and Belden 2002). Mean big sagebrush density
was generally highest in the 0 kg PLS ha–1 grass seeding rate
across historical sampling years; therefore, percent survival
was actually higher at the 16 and 32 kg PLS ha–1 grass
seeding rates.
Big sagebrush densities inside and outside the exclosure
were combined (averaged) in both 2001 and 2002 to compare
to previous years. Although there were no differences in big
sagebrush density among grass seeding rates in 2001 or
2002, there was a consistent decline in big sagebrush density
USDA Forest Service Proceedings RMRS-P-31. 2004
4
3.5
Big sagebrush density (m–2)
Differences (P £ 0.10) in mean big sagebrush density from
permanent quadrats and belt transects, percent big sagebrush plants browsed, mean leader length, and percent utilization were evaluated between grass seeding rates inside and
outside the exclosure using analysis of variance (ANOVA).
Mean separations were evaluated using Tukey’s pairwise
comparison test (experimentwise P £ 0.10) (Krebs 1999).
Inside
Ab
Outside
3
2.5
Aa
2
Aa
Aa
Ba
Aa
1.5
1
0.5
0
0
16
32
Grass seeding rate (kg PLS ha–1)
Figue 3—Mean density of big sagebrush along belt
transects by grass seeding rates inside and outside the
exclosure, North Antelope Coal Mine, Gillette, WY,
2001–2002 (means within a grass seeding rate across
locations with the same uppercase letter are not
significantly different; means within a location across
grass seeding rates with the same lower case letter are
not significantly different, P £ 0.10).
201
Partlow, Olson, Schuman, and Belden
Effects of Wildlife Utilization and Grass Seeding Rates on Big Sagebrush …
120
Big sagebrush density (m–2)
Aa
Aa
Aa
2.5
2
Aa
Inside
Outside
Aa
Aa
Ba
1.5
Ba
1
0.5
0
Percent big sagebrush browsed
3
100
Inside
Outside
Bb
Bb
Bb
80
60
40
Aa Aa
20
Ab
0
June
2001
September
2001
April
2002
September
2002
Ab
June
2001
September
2001
April
2002
Ab
September
2002
Figure 4—Density of big sagebrush along belt
transects by sampling date inside and outside the
exclosure, North Antelope Coal Mine, Gillette, WY,
2001–2002 (means within a sampling date across
locations with the same uppercase letter are not
significantly different; means within a location
across sampling dates with the same lowercase
letter are not significantly different, P £ 0.10).
Figure 5—Percent big sagebrush plants browsed by
sampling date inside and outside the exclosure, North
Antelope Coal Mine, Gillette, WY, 2001–2002 (means
within a sampling date across locations with the same
uppercase letter are not significantly different; means
within a location across sampling dates with the same
lowercase letter are not significantly different, P £ 0.10).
the exclosure compared to outside of the exclosure in April
and September 2002. No differences in density were observed inside the exclosure versus outside of the exclosure in
2001, which was anticipated since the exclosure was not
constructed until June 2001.
Big sagebrush density was positively affected by wildlife
exclusion 10 months (April 2002) after construction of the
exclosure. Big sagebrush density outside the exclosure decreased more rapidly than inside, as evidenced by the number of dead marked plants during the study. Eight marked
sagebrush plants inside the exclosure died compared to 24
plants outside of the exclosure. Mortality outside of the
exclosure, among the marked plants, was 42 percent in the
0 kg PLS ha–1 grass seeding rate, 38 percent in the 16 kg
PLS ha–1 grass seeding rate, and only 21 percent in the 32
kg PLS ha–1 grass seeding rate. Schuman and Belden
(2002) also reported significantly greater sagebrush plant
mortality in the 0 and 16 kg PLS ha–1 compared to the 32 kg
PLS ha–1 grass seeding rate on these plots after 8 years. They
hypothesized that this might be due in part to intraspecific
competition and some sort of protective mechanism of the
grass plants. Owens and Norton (1992) also reported that
basin big sagebrush (Artemisia tridentata ssp. tridentata)
seedlings experienced the greatest mortality in unsheltered
areas of the landscape.
sagebrush browsed decreased inside the exclosure from
June 2001 to September 2002. Browsing inside the exclosure
on the June 2001 sampling date occurred prior to construction of the exclosure about 10 days before the sampling date.
All marked plants outside the exclosure exhibited browsing
regardless of grass seeding rate. Across all grass seeding
rates, April 2002 browsing data suggest that rabbits rather
than big game were the primary browsers of the big sagebrush plants. However, the September 2002 browsing data
indicate that big game and rabbit browsing was nearly
equal. It is important to remember that identifying the
browsing animal is limited to the most recent browser,
because if a plant was browsed by big game and then
browsed by a rabbit the data would only indicate it was
browsed by a rabbit because of the methodology.
Percent Browsed Big Sagebrush Plants
Percent of big sagebrush plants browsed exhibited a
significant interaction between location (inside versus outside exclosure) and sample date (fig. 5). There were no
differences in the degree of big sagebrush plants browsed
inside versus outside the exclosure in June 2001 due to
the recent exclosure construction. However, percent of big
202
Utilization
Big sagebrush leader length exhibited a significant location by grass seeding interaction (fig. 6). Mean leader length
was significantly greater in the 32 kg PLS ha–1 grass seeding
rate compared to the lower grass seeding rates. This response cannot be explained. However, leader lengths were
greater inside of the exclosure compared to outside the
exclosure for all grass seeding rates. There were no differences in leader lengths among grass seeding rates outside of
the exclosure. Big sagebrush leader length was also greater
inside the exclosure for all sample dates (fig. 7). Big sagebrush leader length responded to the protection provided by
the exclosure. Leader length inside the exclosure continued
to increase, while unprotected plants outside the exclosure
exhibited dramatic decreases in leader length due to wildlife
browsing. Big sagebrush survival outside of the exclosure is
threatened by continued intense wildlife utilization.
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Wildlife Utilization and Grass Seeding Rates on Bg Sagebrush …
Partlow, Olson, Schuman, and Belden
60
60
Ba
Inside
Outside
Aab
Ab
Aa
40
30
20
Ba
Ba
Summer
50
Ba
10
Utilization (percent)
Mean leader length (mm)
50
Winter
Bb
40
30
20
Ac
Aa
Aa
Aa
10
0
0
16
0
32
0
Grass seeding rates (kg PLS ha–1)
16
32
Grass seeding rates (kg PLS ha–1)
Figure 6—Mean leader length of marked big sagebrush by grass seeding rates inside and outside the
exclosure, North Antelope Coal Mine, Gillette, WY,
2001–2002 (means within a grass seeding rate across
locations with the same uppercase letter are not
significantly different; means within a location across
grass seeding rates with the same lower case letter
are not significantly different, P £ 0.10).
Figure 8—Seasonal percent utilization of marked big
sagebrush plants by grass seeding rates outside the
exclosure, North Antelope Coal Mine, Gillette, WY,
2001–2002 (means within a grass seeding rate across
season with the same uppercase letter are not
significantly different; means within a season across
grass seeding rates with the same lower case letter are
not significantly different, P £ 0.10).
Seasonal utilization of marked sagebrush plants outside
the exclosure exhibited a significant interaction between
grass seeding rate and sample period (fig. 8). Summer
wildlife utilization was consistent across grass seeding rates.
However, winter utilization was significantly lower in the 32
kg PLS ha–1 grass seeding rate compared to both the 16 and
0 kg PLS ha–1 grass seeding rates. There was also significantly greater wildlife utilization in the winter compared to
the summer period in the 0 and 16 kg PLS ha–1 grass seeding
rates. The lower wildlife utilization during the winter months
as grass seeding rate increased agrees with the sagebrush
survival data presented by Schuman and Belden (2002)
after 8 years, and also agrees with the survival and mortality
data presented here 10 years after establishment (fig. 9). It
appears that several things may be influencing the response
sagebrush are exhibiting on this site. Austin and others
(1994) evaluated the effects of deer and horse browsing on
transplanted Wyoming big sagebrush survival. They found
that plantings resulting in densities of 0.08 and 0.44 sagebrush plants m–2 were not influenced by the animals. However, the plant densities at our study site were severalfold
higher and may have resulted in intraspecific competition
for space, water, and nutrients. Owens and Norton (1994)
found that sagebrush seedlings sheltered by other plants
experienced less mortality than those growing in unprotected spaces. Based on their findings and those of Schuman
and Belden (2002) it appears there are likely several factors
affecting sagebrush survival on this reclaimed mined land.
It is obvious from the significantly higher survival, lower
mortality, and reduced wildlife utilization (fig. 9) of big
sagebrush at the higher grass seeding rates that this is not
a random response even though we cannot fully explain it.
Greater winter utilization was anticipated because browsing preference by big game (Craven 1983b; Schemnitz 1982;
Welch and others 1981) and rabbits (Anderson and Shumar
1986; Craven 1983a; Knight 1982) intensify during the
winter. Big sagebrush is a “starvation food” or necessary
dietary component of winter months when nothing else is
available (Welch and others 1981). Partial or complete defoliation of sagebrush leaders will not adversely affect growth,
70
Mean leader length (mm)
60
Ac
Inside
Outside
50
40
Ab
Aa
Aa
30
20
Bab
Ba
Bb
Bc
10
0
June
2001
September
2001
April
2002
Figure 7—Mean leader length of marked big sagebrush
plants by sampling period inside and outside the
exclosure, North Antelope Coal Mine, Gillette, WY,
2001–2002 (means within a sampling date across
locations with the same uppercase letter are not
significantly different; means within a location across
sampling dates with the same lowercase letter are not
significantly different, P £ 0.10).
USDA Forest Service Proceedings RMRS-P-31. 2004
September
2002
203
Partlow, Olson, Schuman, and Belden
Effects of Wildlife Utilization and Grass Seeding Rates on Big Sagebrush …
80
% Winter Utilization
% Survival
70
% Mortality
60
50
40
30
20
10
0
0
16
32
Grass seeding rate (kg PLS ha–1)
Figure 9—Relationship between percent winter
utilization (2002), survival, and mortality of big
sagebrush plants, North Antelope Coal Mine, Gillette,
WY, 1994–2002.
vigor, and survival if leaf primordia and twigs are undamaged (Kelsey 1984). However, defoliation in our study was
much more severe with considerable twig and primordia
damage. Wildlife contributed to the death of about 33 percent of the marked sagebrush plants outside of the exclosure
within 15 months.
For reclamation sites to provide adequate wildlife habitat,
big sagebrush must be successfully established and maintained. Big sagebrush is included in the reclamation seed
mix to satisfy State and Federal regulations for shrub reestablishment and land use. However, the intended land use
(wildlife habitat) seems to be the very reason for threatened
big sagebrush survival at this site. Our data indicate that
high wildlife densities are impeding the successful longterm establishment and growth of the big sagebrush component of the reclaimed site. Both wildlife management and
habitat manipulation may be necessary for successful reclamation of this site. Without proper wildlife and habitat
management, big sagebrush densities could decline to less
than 1 plant m–2 on these reclaimed areas, and not meet the
required density for bond release (Wyoming Department of
Environmental Quality, Land Quality Division 1996).
Management Implications ________
Because intense wildlife utilization has been shown to
reduce big sagebrush survival on some mine sites, reclamation specialists should consider management practices to
reduce wildlife herbivory impacts. Habitats on adjacent,
native rangeland may be improved to attract wildlife away
from reclamation sites, possibly enhancing big sagebrush
survival. Prescribed burning, mechanical practices (mowing, roto-beating), and other treatments of adjacent native
big sagebrush-dominated rangeland that increase herbaceous plant production, improve forage quality, and enhance
plant diversity may help distribute wildlife. Interseeding
perennial grasses and forbs, combined with management of
overmature big sagebrush stands, could improve adjacent
204
native landscape condition and improve big sagebrush survival on reclaimed areas by increasing wildlife distribution.
Improving wildlife distribution and enhancing rangeland
forage quality may not be enough. Therefore, wildlife population management may also be necessary at some mines.
Wildlife are attracted to reclamation areas for foraging on
freshly seeded plant species, including big sagebrush seedlings. In addition, wildlife are provided protection from
human disturbance by restricted public access and prohibition of hunting within the mine permit area. Wildlife populations might be better managed on theses reclaimed areas
if mining companies would consider allowing limited harvesting (hunting) by methods compatible with the mine
environment and considerations.
Evaluating the impacts of wildlife browsing on big sagebrush survival is necessary for all mines trying to reestablish big sagebrush and other shrub species. This study
has additional value as a demonstration site for illustrating
long-term differences between browsed (outside exclosure)
and unbrowsed (inside exclosure) big sagebrush years after
project completion. Recommendations to reduce wildlife
browsing impacts on big sagebrush, as a byproduct of this
research, will hopefully result in more successful reclamation by mining companies and provide improved wildlife
habitat.
Acknowledgments _____________
The authors would like to acknowledge the assistance of
Matt Mortenson, Cliff Bowen, Lachlan Ingram, Krissie
Peterson, Margaret Sharp, and Kelli Sutpfin in field data
collection. We would also like to acknowledge Powder River
Coal Co., North Antelope/Rochelle Complex, Peabody Energy Co., Gillette, WY, and Abandoned Coal Mine Land
Research Program, Office of Research, University of Wyoming, and the Wyoming Department of Environmental
Quality, Abandoned Mine Land Division, Cheyenne, WY, for
partial funding of this research.
References ____________________
Anderson, J. E.; Schumar, M. L. 1986. Impacts of black-tailed
jackrabbits at peak population densities on sagebrush-steppe
vegetation. Journal of Range Management. 39: 152–156.
Beetle, A. A. 1960. A study of sagebrush: the section tridentata of
Artemisia. Bull. 368. Laramie: Wyoming Agricultural Experiment Station. 83 p.
Cockrell, J. R.; Schuman, G. E.; Booth, D. T. 1995. Evaluation of
cultural methods for establishing Wyoming big sagebrush on
mined lands. In: Schuman, G. E.; Vance. G. F., eds. Decades later:
a time for reassessment. Proceedings of 12th annual meeting,
American Society for Surface Mining and Reclamation; 1995
June 5–8; Gillette, WY. Princeton, WV: 784–795.
Craven, S. R. 1983a. Cottontail rabbits. In: Timm, R. M., ed.
Prevention and control of wildlife damage. Lincoln, NE: Great
Plains Agricultural Council Wildlife Resources Committee and
University of Nebraska Cooperative Extension Service: D69–D74.
Craven, S. R. 1983b. Deer. In: Timm, R. M., ed. Prevention and
control of wildlife damage. Lincoln, NE: Great Plains Agricultural Council Wildlife Resources Committee and University of
Nebraska Cooperative Extension Service: D23–D34.
DePuit, E. J. 1988. Shrub establishment on drastically disturbed
lands. In: Fisser, H. G., ed. Proceedings of the seventeenth
USDA Forest Service Proceedings RMRS-P-31. 2004
Effects of Wildlife Utilization and Grass Seeding Rates on Bg Sagebrush …
Wyoming shrub ecology workshop; 1988 June 21–22; Jackson,
WY. Laramie: University of Wyoming: 55–60.
Glassey, T. W.; Dunnewald, T. J.; Brock, J.; Irving, H. H.; Tippetts, N.;
Rohrer, C. 1955. Campbell County soil survey, Wyoming. Soil
Conservation Service, Number 22. Washington, DC: U.S. Department of Agriculture, United States Government Printing Office.
Kelsey, R. G. 1984. Foliage biomass and crude terpenoid productivity of big sagebrush. In: McArthur, E. D.; Welch, B. L., comps.
Proceedings: symposium on the biology of Artemisia and
Chrysothamnus; 1984 July 9–13; Provo, UT. Gen. Tech. Rep. INT200. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Inermountain Research Station: 375–388.
Knight. J. E. 1983. Jackrabbits. In: Timm, R. M., ed. Prevention and
control of wildlife damage. Lincoln, NE: Great Plains Agricultural Council Wildlife Resources Committee and University of
Nebraska Cooperative Extension Service: D75–D80.
Krebs, C. J. 1999. Ecological methodology. New York: Harper and
Row, Publishers.
McNaughton, S. J. 1983. Compensatory plant growth as a response
to herbivory. Oikos. 40: 329–336.
Schemnitz, S. D. 1983. Pronghorn antelope. In: Timm, R. M., ed.
Prevention and control of wildlife damage. Lincoln, NE: Great
Plains Agricultural Council Wildlife Resources Committee and
University of Nebraska Cooperative Extension Service: D1–D4.
Schuman, G. E.; Booth, D. T.; Cockrell, J. R. 1998. Cultural methods
for establishing Wyoming big sagebrush on mined lands. Journal
of Range Management. 51: 223–230.
USDA Forest Service Proceedings RMRS-P-31. 2004
Partlow, Olson, Schuman, and Belden
Schuman, G. E.; Belden, S. E. 2002. Long term survival of direct
seeded Wyoming big sagebrush on a reclaimed mine site. Arid
Land Research and Management.16: 309–317.
Stahl, P. D.; Williams, S. E.; Christensen, M. 1988. Efficacy of native
vesicular-arbuscular mycorrhizal fungi under severe soil disturbance. New Phytology. 110: 347–354.
Wambolt, C. L. 1996. Mule deer and elk foraging preference for four
sagebrush taxa. Journal of Range Management. 49(6): 499–503.
Welch, D. L.; McArthur, E. D.; Davis, J. N. 1981. Differential preference of wintering mule deer for accessions of big sagebrush and
for black sagebrush. Journal of Range Management. 32: 467-469.
Western Water Consultants and Bureau of Land Management.
1998. Environmental impact statement for the Powder River
Coal Lease Application and the Thundercloud Coal Lease Application. Sheridan, WY: Western Water Consultants; and Casper,
WY: U.S. Department of the Interior, Bureau of Land Management.
White, J. A.; Munn, L. C.; Williams, S. E. 1989. Edaphic and reclamation aspects of vesicular-arbuscular mycorrhizae in Wyoming
Red Desert soils. Soil Science Society of America Journal. 53: 86–90.
Wyoming Department of Environmental Quality, Land Quality Division. 1996. Coal rules and regulations. Chapter 4, Appendix A.
Cheyenne: State of Wyoming.
Young, J. F.; Singleton, P. C. 1977. Wyoming general soil map.
Research Journal 117. Laramie: University of Wyoming, Wyoming Agricultural Experiment Station.
205
Field Trips
Spirdiluv
207
Illustration on the reverse side is by Carolyn Crawford. In: Fertig, W.; Refsdal, C.; Whipple, J. 1994. Wyoming rare plant field guide. Cheyenne: Wyoming
Rare Plant Technical Committee. Available at: http://www.npwrc.usgs.gov/resource/distr/others/wyplant/species.htm (Accessed 12/18/03).
Research at the High Plains
Grasslands Research Station
D. T. Booth
J. D. Derner
Abstract: During the 12th Wildland Shrub Symposium, August 12
to 16, 2002, in Laramie, WY, field trip participants visited the High
Plains Grasslands Research Station in Cheyenne. The Station is
operated by the U.S. Department of Agriculture, Agricultural Research Service on land owned by the city of Cheyenne. The tour
included a chance to visit the historic setting, tour facilities, and
listen to presentations by Station scientists and cooperating investigators. Cooperators include the University of Wyoming and Colorado State University; U.S. Department of the Interior, Bureau of
Land Management, Wyoming State Office; and the U.S. Department of Agriculture, Natural Resources Conservation Service, Grazing Lands Team.
Introduction ____________________
The High Plains Grasslands Research Station (HPGRS) is
headquarters for the U.S. Department of Agriculture, Agricultural Research Service, Rangeland Resources Research
Unit (RRRU). Besides the Cheyenne Station, the unit has
field, office, and laboratory resources at the Central Plains
Experimental Range (CPER), northeast of Nunn, CO, and
offices and laboratories at the Crops Research Laboratory in
Fort Collins, CO. The two field locations (HPGRS and CPER)
are separated by less than 25 miles but have contrasting
vegetation that greatly enhances research opportunities
(table 1). The HPGRS is located at the southern end of the
northern mixed-grass prairie, while the CPER is located at
the northern edge of the shortgrass prairie. The CPER is also
a part of worldwide Long-Term Ecological Research (LTER;
http://lternet.edu). The research unit has a strong plantanimal-soil focus with a breadth of scientific expertise that
has undergone some changes with two scientist retirements
and two new scientists hired in 2002 (table 2).
History
In 1928, the 70th U.S. Congress authorized and directed
the U.S. Department of Agriculture to establish an experimental station near Cheyenne, WY (Schuman 1989). To that
end, 2,139 acres (a third of which was irrigable) were leased
from the city of Cheyenne for 199 years at $1 per year. The
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004s. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
D. T. Booth and J. D. Derner are Rangeland Scientists, U.S. Department
of Agriculture, Agricultural Research Service, High Plains Grasslands Research Station, 8408 Hildreth Road, Cheyenne, WY 82009, U.S.A., e-mail:
tbooth@lamar.colostate.edu
USDA Forest Service Proceedings RMRS-P-31. 2004
station was named the Central Great Plains Field Station
and was meant to complement the Northern Great Plains
Field Station at Mandan, ND, and the Southern Great
Plains Field Station at Woodward, OK. The first Superintendent at Cheyenne was Robert Wilson (1928 to 1930), and
construction of the buildings began in June 1928. In 1930,
A. C. Hildreth (1930 to 1959) became the Superintendent,
the Station was renamed the Cheyenne Horticultural Field
Station, and the first plantings of trees, shrubs, fruits, and
vegetables were made.
The following year construction of the initial buildings
was completed, and the first of numerous shelterbelts and
landscape plantings were made on Federal, State, and local
cooperating government lands. During the ensuing years of
shelterbelt research, an additional 323 shelterbelts were
planted on cooperating private farms in Wyoming, Colorado,
Nebraska, Kansas, Utah, and South Dakota. This planting
effort was facilitated by a Civilian Conservation Corp camp
of 200 men opened on the Station in 1935. In its 7-year
history the Corp constructed roads, water and septic systems, and over 2 miles of concrete-lined irrigation ditches, in
addition to planting thousands of trees and shrubs.
Besides the shelterbelt research, over 1,300 varieties of
fruit trees and 300 varieties of small fruits like raspberries,
currants, gooseberries, and strawberries were included in
varietal tests seeking strains adapted to the cold and drought
of the High Plains. Many native strawberries (4,200 collections) were obtained from the Rocky Mountain area from
Montana to New Mexico. These collections later led to the
release of several strawberry varieties including Radiance,
Ogallala, and Fort Laramie.
The Station’s mission was broadened in 1936 to include
forage crop research, and over 200 species and varieties of
grass were established to evaluate their yield potential,
hardiness, and nutritive value. The first horticultural findings from Station research were published that year, and in
1940 publication of the first Station grass research followed.
In 1942, Hildreth and others were temporarily reassigned
to California to study the guayule plant as an alternative
rubber source. Later Hildreth helped establish an agricultural experiment station in Afghanistan as part of his official
duties. He retired in 1959, and L. A. Schaal became Superintendent in 1960. Schaal initiated a potato research program in 1961 in cooperation with the USDA station in
Greeley, CO, and the University of Wyoming. In 1962 the
Station published fruit-tree test results covering 28 years
and 1,200 apple, 300 plum, 50 cherry, and 40 pear varieties.
Gene Howard initiated carnation and chrysanthemum research after becoming superintendent in 1964. The Cheyenne Hardy Mums were named Cheyenne’s official flower in
209
Booth
Research at the High Plains Grassland Research Station
Table 1—Characteristics of research field locations.
Characteristics
High Plains Grasslands
Research Station
Cheyenne, WY
Central Plains
Experimental Range
Nunn, CO
Size
Annual precipitation
Rangeland type
Annual production
2,870 acres
14.6 inches
Mixed-grass prairie
1,380 pounds per acre
15,500 acres
12.8 inches
Shortgrass prairie
625 pounds per acre
Table 2—High Plains Grasslands Research Station scientists and field of expertise.
Name
Jack Morgan
Gerald Schuman
Terrance Booth
Jean Reeder
Justin Dernera
Dana Blumenthala
Richard Hart, retired
Gary Frasier, retired
a
Title
Plant Physiologist
Soil Scientist
Rangeland Scientist
Soil Scientist
Rangeland Scientist
Weed Ecologist
Agronomist
Hydrologist
Carbon and water fluxes, global climate change
Carbon sequestration, reclamation of degraded lands
Revegetation, remote sensing for rangeland monitoring
Nutrient cycling, belowground processes, carbon sequestration
Rangeland monitoring and health, grazing management
Invasive species, global change
Grazing management
Rangeland hydrology
Newly hired.
1970. Then in 1974 Howard had the task of terminating the
horticultural research program—fruits, flowers, vegetables,
and shelterbelts—and of developing a program directed
toward grazing management, water conservation, and minedland reclamation. As part of the program redirection the
Station was renamed the High Plains Grasslands Research
Station and several new staff members were added.
Richard Hart became Location Leader (Superintendent)
in l976, and the first papers derived from the Station’s new
grassland research mission were published. Additional land
was acquired from F. E. Warren Air Force Base, enabling
further development of livestock grazing research. In 1981,
Gerald Schuman became the Location Leader and continued
in that capacity until 1998. During that time grazing management and mined-land reclamation were the predominant research activities, and research findings contributed
to improved stocking rate recommendations for rangelands,
and to increasingly successful techniques for mined land
reclamation. Jack Morgan assumed leadership responsibilities in 1998 and implemented new research with a focus on
carbon sequestration, rangeland monitoring and health,
management strategies for mitigation of global climate
change, and invasive species.
New Changes to Meet the
Challenges of the Twenty-First
Century ________________________
Rangeland monitoring and health, invasive species, and
improved land management strategies will be the primary
research foci for the HPGRS in the early part of the new
210
Research
century. Research will continue to target systems-level
questions that are pertinent to both private and public land
issues (grazing, revegetation, weed management), but these
practices will be addressed in the context of the ecological
system and sustaining rangeland health. While research
will continue at the HPGRS and CPER field sites, Station
scientists will expand their research into the sagebrush
steppe vegetation that is prominent throughout much of
Wyoming. New research projects targeted to address public
lands questions pertinent to the sagebrush steppe vegetation include determining above- and belowground carbon
budgets following changes in land management and utilizing very large scale aerial (VLSA) imagery as a method to
assess and monitor vegetation change.
The High Plains Grasslands Research Station has a long
history of providing practical information for land managers
and of reacting to the needs of local, State, and national
customers. To this end, HPGRS scientists will continue to
investigate innovative approaches to land management that
combine emerging technologies and customer-oriented products. Our mission is to learn how plants, soil, water, atmosphere, weather, and management interact on grazing lands,
and to use the resulting knowledge to promote sustainable
and profitable management systems producing desired goods
and services.
Reference ______________________
Schuman, G. E. 1989. Cheyenne Horticultural Field Station/High
Plains Grasslands Research Station. In: Field, S. L., ed. History
of Cheyenne, Wyoming-Laramie County. Vol. 2. Dallas, TX:
Curtis Media Corporation: 55–56.
USDA Forest Service Proceedings RMRS-P-31. 2004
Research at the High Plains Grassland Research Station
Booth
Appendix A: Agenda for Wildland Shrub Symposium Field Trip to the High
Plains Grasslands Research Station__________________________________
Wednesday, August 14, 2002
8:00 a.m.
9:00
9:10 & 10:00
11:20
12:00
1:30 p.m.
2:00
3:00
Leave Laramie and travel to the High Plains Grasslands Research Station via Happy Jack Road
Welcome to the High Plains Grasslands Research Station by Jack Morgan, Research Leader
Oral presentation in main office building.
Plant and animal responses to grazing systems at the High Plains Grasslands Research Station by
Justin Derner and Jim Waggoner
Poster presentations at picnic area (presenters will be at their posters until 11:20)
Elevated CO2 enhances production but reduces forage quality in perennial grasses of the
shortgrass steppe - Jack Morgan
Influence of livestock grazing on carbon sequestration in a semiarid mixed-grass and
short-grass rangelands - Jean Reeder and Gerald Schuman
Effect of interseeded Falcata alfalfa on carbon sequestration and production of a mixed-grass
rangeland - Gerald Schuman
Transplantable container for rangeland shrub revegetation - Terrance Booth
Presentation at the airplane:
Rangeland monitoring using very large scale imagery - Terrance Booth, Joe Nance, and Sam Cox
Break (visit labs, see Seed Growth Robot-2 and Wyoming big sagebrush seed orchard, or visit with staff)
Leave the High Plains Grasslands Research Station for Cheyenne’s botanical garden and Lion’s Park.
A sack lunch will be served at Lion’s Park followed by a tour of the botanical gardens or just enjoy the park.
Leave Lion’s Park for Terry Bison Ranch
Arrive Terry Bison Ranch and visit about ranch forage resources and grazing management.
Return to Laramie via I-80
USDA Forest Service Proceedings RMRS-P-31. 2004
211
Soils and Vegetation of the
Snowy Range, Southeastern
Wyoming Wildland Shrub
Symposium Field Trip: Laramie
to Woods Landing, Riverside,
Libby Flats, and Centennial,
Wednesday, August 14, 2002
Larry Munn
C. Lynn Kinter
Laramie, Wyoming, was established as a railroad construction town in May, 1868. The transcontinental railroad
(Union Pacific) came across the Rockies in the late 1860s,
across the then unsettled landscape that now contains
Cheyenne and Laramie, after a southern route through
Colorado was rejected because of the terrain. A relic of the
Tertiary Ogallala formation called the “Gangplank” west of
Cheyenne provided an access ramp across the mountains;
this was the first place north of Denver where it was feasible
to build the railroad. It was also the first place west of
Missouri that railroad ties could be cut from the surrounding
forests, and railroaders timbered for ties extensively in the
Laramie and Snowy Ranges (about 3,000,000 ties were cut
by 1870). The “tie hacking” continued until World War II.
“Tie hacks” would cut mid-sized trees of 12 to 14 inches (30
to 35 cm) diameter in the winter, cutting flats on two
opposing sides of the tree with a broad axe and piling 8-ft
(2.4-m) lengths into creeks all winter, behind a temporary
dam. In spring, when runoff waters were high, the dams
were dynamited, and they had a log run! The fishery biologists are still trying to restore fish habitat in some of the
drainages in the Snowy Range where channels were extensively gouged by these log runs.
Today, you can still see standing stumps from tie-hack
activity in the Laramie and Snowy Ranges—often they are
3 to 4 ft (0.9 to 1.2 m) tall because the hacks were cutting in
deep snow. These forests provided ties from Laramie to
Rawlins and Rock Springs, 200 miles away, at which point
ties were cut from the forests to the south of the track in the
Uinta Range. At the peak of the steam engine period, the
In: Hild, Ann L.; Shaw, Nancy L.; Meyer, Susan E.; Booth, D. Terrance;
McArthur, E. Durant, comps. 2004. Seed and soil dynamics in shrubland
ecosystems: proceedings; 2002 August 12–16; Laramie, WY. Proceedings
RMRS-P-31. Ogden, UT: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station.
Larry Munn is Professor and Soil Scientist, Department of Renewable
Resources, University of Wyoming, Laramie, WY; e-mail: LCMunn@uwyo.edu.
C. Lynn Kinter is Botanist and Doctoral Candidate, School of Biological
Sciences, Washington State University.
212
trains also used 2 million gallons of water each day from the
Laramie River. Stops along the track where coal and water
were available developed into towns, for example, Rawlins
and Rock Springs. We left town heading southwest via
Highway 230. Elevation at Laramie is 7,170 ft (2,185 m), and
precipitation is 10.5 inches (26 cm), most of which falls in
March through July.
Stop 1: University of Wyoming, Monolith Ranch by
the fishing access to a Laramie River site with greasewood
(Sarcobatus vermiculatus). This is an area that was farmed
at the turn of the last century. In 1906, a grain elevator
capable of loading 50,000 bushels onto railroad cars daily
was built in Laramie. By 1920, failed farms were being
consolidated into ranches. Greasewood, a highly salt-tolerant shrub, is increasing here between the Pioneer Canal
(dug in 1890, visible along the hillside to the west and north)
and the Laramie River to the southeast. The greasewood is
expanding from canal seepage water that for the last hundred years has permeated the unlined canal into the underlying Niobrara formation (a Cretaceous age marine shale
and sandstone). Other salt-tolerant plants include alkali
sacaton (Sporobolus airoides) and saltbush (Atriplex sp.).
Soils here are apparently increasing in salinity over time as
a result of an accumulation of very soluble salts from the
canal seepage water. The soils on this geomorphic surface
are Aridisols (Calcids with frigid temperature regimes),
which typically have calcic horizons and significant amounts
of gypsum in the subsoil.
Stop 2: Fox Creek Road. Here we are on an alluvial fan
extending from a canyon. The soils have a fairly thick
profile; however, they are not well developed (no B horizons), but do have a mollic epipedon. Elevation at the site
is 7,800 ft (2,377 m), and annual precipitation is approximately 16 inches (40 cm). There is a pronounced aspect
effect here, between north- and south-facing slopes. Soils
on the landscape are Mollisols (Haplustolls) on the drier
south slopes; on the wetter, north-facing slopes the soils are
Inseptisols (Dystrocryepts). There is also a pronounced
aspect difference between north- and south-facing slopes in
USDA Forest Service Proceedings RMRS-P-31. 2004
Soils and Vegetation of the Snowy Range, Southeastern Wyoming Wildland ...
Munn and Kinter
terms of the trees present: lodgepole pine (Pinus contorta),
subalpine fir (Abies lasiocarpa), Englemann spruce (Picea
engelmannii), and Douglas-fir (Pseudotsuga menziesii) on
north-facing slopes; ponderosa pine (Pinus ponderosa),
limber pine (Pinus flexilis), and Rocky Mountain juniper
(Juniperus scopulorum) on the south-facing slopes. Dry
open sites are dominated by mountain big sagebrush (Artemisia tridentata var. vaseyana), common rabbitbrush
(Chrysothamnus nauseosus), and Idaho fescue (Festuca
idahoensis). Narrowleaf cottonwood (Populus angustifolia),
balsam poplar (Populus balsamifera), mountain alder (Alnus
incana), willows (Salix spp.) and choke-cherry (Prunus
virginiana) grow in the riparian area along Fox Creek.
Chokecherries are an important food supplement for black
bears; there are approximately 100 bears in the Snowy
Range. This is an area that is used heavily by deer and elk
as winter range, so you may see signs prohibiting offroad
travel during the winter months.
As we begin to turn north into Wyoming, on the slopes to
the south of the road, remains a scar from a fairly large fire
where an extensive stand of conifers burned about 10 to 15
years ago. That is a common story here in the Northern
Rockies where through fire control we have allowed canopy
closure in the forest on many sites that were probably more
open historically. The first fire control plan was established
on the Medicine Bow National Forest in 1912. These sites
now are contributing to the fire problem in the West because
current fires are typically hotter and extend over much
larger areas than those that occurred prior to fire suppression.
On the north side of the road, there is an old copper mine.
Though much more hard rock mining occurred in Colorado
than in Wyoming, the Snowy Range does contain a series of
gold mines and copper mines. Over in the Sierra Madres to
the west of us, there was a copper mine with a smelter in
Encampment, a town we will come to later in the trip.
Along the Highway: As we travel to higher elevations
near the southern end of the Snowy Range, we drive past the
road to the old sawmill settlement at Fox Park. At one time,
the Forest Service had a small workstation there. The
sawmill is closed now, and the area is largely private cabins.
The Fox Park area was extensively burned 150 to 200
years ago. It is a large area of fairly uniform lodgepole pine
forest interspersed with moist shrub meadows along streams.
In the coniferous forest, understory shrubs such as whortleberry (Vaccinium scoparium) and bearberry (Arctostaphylos
uva-ursi) provide important forage for wildlife. Soils here
are primarily Alfisols (Haplocryalfs), and in the meadows
are wet Mollisols (Cryaquolls) or wet Inceptisols (Cryaquepts).
This area of the Snowy Range was not glaciated, and the
soils are relatively well developed (thicker Bt horizons) and
contain more clay than soils on the younger, glaciated
landscapes we will travel over later in the trip. Consequently, the water-holding capacity of these soils is greater,
and the forests are more productive. Underlying geology
here at the southern end of the Snowys is a complex of
granitic, felsic, and hornblende gneiss, as well as amphibolite.
As we continue past Mountain Home we see several
meadows with Wyoming big sagebrush (Artemisia tridentata
var. wyomingensis). This is important habitat for wildlife
that often finds limited food under the conifers. These areas
also provide nesting habitat for numerous species of birds.
Next, the highway crosses briefly into northern Colorado.
Here, we find larger openings of sagebrush grassland. You
might notice on the surrounding hills that in the past, the
Forest Service allowed rather large clearcuts. These are
currently revegetated in aspen (Populus tremuloides) or
young conifers, but are still clearly identifiable by the size of
the trees. Today, logging cuts are more considerate of wildlife habitat needs, are generally smaller, and are done to
create irregular contours and are thus less apparent in the
early regrowth phase.
Our path has followed a former railroad track bed; the
rails have now been removed. This railroad went from
Laramie to North Park Colorado and was built in about 1910
to haul wood products from the Fox Park area and coal from
a mine in North Park, Colorado. That coal mine has been
shut down for almost 20 years now and the rails and ties
were sold for scrap.
USDA Forest Service Proceedings RMRS-P-31. 2004
As we pass the North Platte River we can see that it
has fairly low flow this year, reflecting the drought we have
been experiencing here in Wyoming for the past 3 to 5 years.
As we drive along the Platte River Valley toward Saratoga,
we see a landscape that is underlain by Tertiary age geologic
materials (Upper Miocene, approximately 20 million years
old). The geologic formation is the North Park Formation.
There are small areas of older Eocene deposits along the
southwest fringe of the basin. The small hills to the east
along the highway are granite, and you may get some feeling
of the filling of the basin with Tertiary valley fill materials.
The geologic materials we are driving over were eroded from
the top of the adjacent mountains as they were uplifted
during the Tertiary. At several times during the Tertiary,
the intermountain basins were largely filled with alluvium
eroded off the rising mountains. Then, during periods of
quietus in mountain uplift, the streams eroded much of the
sediment from the basins toward the Gulf of Mexico. We are
now apparently in a period when the uplift is relatively quiet
and the basins are being exhumed slowly by geologic erosion.
The light-colored sediments that show on hillsides along the
road are exposures of the Tertiary valley fill.
We are now in a rolling vista of Wyoming big sagebrush
(Artemisia tridentata var. wyomingensis) and several species of wheatgrass (Agropyron spp.). We also see areas where
the sagebrush was sprayed for control—as shown by the
delineations across the landscape.
Stop 3: We turn off at the Six Mile Road access to the
National Forest. After traversing an area of private land, we
reach a fishing access by the Platte River that is very
important winter range for elk, deer, and bighorn sheep. The
Wyoming Game and Fish Department estimates that there
are approximately 6,500 elk and 250 bighorn sheep in the
Snowy Range at present. About 600 elk and many of the
sheep winter in this area.
The elevation at the Six Mile access to the Platte River is
7,500 ft (2,286 m). and the precipitation is about 16 inches
(40 cm). This area was burned about 6 years ago for wildlife
habitat improvement. Most burning is done in April to catch
a window when the snow has melted from the sagebrush but
is still present in the trees. This timing also prevents
damage to root systems of shrubs like bitterbrush (Purshia
tridentata), which sprout from the roots if the fires are not
too hot. The road was used as a firebreak. Below the road and
213
Munn and Kinter
Soils and Vegetation of the Snowy Range, Southeastern Wyoming Wildland ...
to the north up the slope you can see an area that was
burned; the south side of the road represents the unburned
condition. The burned area appears to have good regrowth of
bitterbrush and shrubby cinquefoil (Potentilla fruticosa)
and good recruitment of sagebrush seedlings. The soils here
are Mollisols over the rolling topography with Inceptisols,
and Entisols (Lithic Cryorthents) on the steeper slopes
where the rocks are close to the surface.
quartzite boulders derived from the crest of the Snowy
Range. You can see the peaks in the distance by looking up
the highway. Soils in the willow bottom are primarily wet
Inceptisols (Histic Cryaquepts), which have a thin layer of
organic material over stratified alluvium. The beaver ponds
along the creek act as important sediment traps and slow the
runoff of snowmelt waters through the summer months.
Generally Histosols are only found in very high water table
areas in the riparian zones, and in kettle holes in the
Pinedale moraine.
Along the highway, driving north from the Six Mile
access road, the areas without sagebrush were cleared for
hay or wheat production. Much of the acreage formerly
farmed with wheat in Wyoming has now been abandoned;
the climate is too dry and cold, and the Conservation Reserve
Program provided a good incentive for many farmers to
revegetate back into rangeland.
Stop 4: Forest Service campground at Ryan Park.
This area was the site of a Civilian Conservation Corps camp
in the 1930s and was a prisoner of war camp during World
War II. There was a rather short ski slope here also.
Medicine Bow National Forest via Highway 130. As
we follow the creek you may see very telltale signs of
glaciation. Ice collected on the north end of the Snowy Range
on numerous occasions during the Pleistocene; and glaciers
moved from the top of the range to the north, east, and west
of Medicine Bow Peak. The ice carried glacial till across the
landscape and down the major drainages. We are now
driving along a moraine of Pinedale age (Late Wisconsin age;
approximately 25,000 to 12,000 years before present). You
can see the boulders imbedded in fine material jutting out of
the roadcuts. Most of the soils on the moraines are Inceptisols
because of their relatively young age, given the cool climate
under the forest. Blocks of white rock are very distinctive in
the till; they are the Medicine Peak quartzite from the
highest crest of the Snowy Range. We will eventually drive
beside the cliffs from which these rocks are derived; the rock
was extracted by the ice and carried down the slopes in the
French Creek watershed. The quartzite represents an old
beach (2 billion years old!) and provides a good way to trace
glacial activity from the north part of the range. The glaciers
moved north, toward Elk Mountain; west, via French creek
where we travel along the stream; and eastward, via Libby
Creek toward the town of Centennial. French Creek was
named for a family of Arcadians, the De La Sol family, who
came here to prospect for gold. They did establish a productive claim, but the family eventually sold the mine and
moved further north in Wyoming where they bought and
leased a lot of land to create what is now the largest ranch in
the State—the Sun Ranch. Another creek in the area, Pelton
Creek, was named after a man who came to the area
originally as a guest of the Wyoming Territorial Prison, but
who later became a respected Laramie citizen and businessman (after his release).
Stop 5: Forest Service Riparian Area Walkway. The
elevation here is 8,700 ft (2,652 m), and the precipitation is
approximately 26 inches (65 cm). The riparian system here
is dominated by willows (Salix spp.), sedges (Carex aquatilis,
Carex microptera, and others), tufted hairgrass (Deschampsia
cespitosa), and bluejoint (Calamagrostis canadensis). Across
the road to the north you can see a good exposure of the
moraine; the till contains, again, the white Medicine Peak
214
Stop 6: At the Forest Service observatory on top of the
summit. Elevation here is 10,845 ft (3,305 m). The ice here
was perhaps 800 to 1,000 ft (244 to 305 m) thick during Bull
Lake glaciation (corresponding to the Illinoian glaciation of
the mid-continent 175,000 to 140,000 years ago) and again
later during the Late Pinedale glaciation (25,000 to 12,000
years ago). There were smaller glaciers several thousand
years ago during the Neoglaciation, which started about
4,000 years before present. Ice accumulation during the
Neoglaciation was simply a major sheet of ice that covered
the north and east slopes of the crest. Blocks of rock, loosened
from the exposed crest by frost action, slid along the ice sheet
and formed pro-talus ramparts that are now several hundred yards from the rock wall—thus indicating the dimensions of the ice sheet. The landscape contains Nash dolomite
as the underlying bedrock and quartzite blocks deposited in
a thin smear of till across this landscape. The inclusion of
blocks of the Nash dolomite in the till serves to buffer the pH
of the soils developed here, and this effect carries all the way
down the valleys on either side of the Range.
Medicine Bow Peak at 12,005 ft (3,661 m) is the highest
peak in the Snowy Range. The dark band that parallels the
ridgeline is a mafic intrusion that is much more easily
weathered than the white quartzite. On October 6, 1955, a
four-engine commercial airliner crashed into the peak at
about the mafic intrusion, killing all 36 people on board.
Several engines and other parts of the plane are still present
at the base of the cliff, and in a few more years they will be
protected by the Antiquities Act; now they are just junk!
Krummholtz trees, many of which are Engelmann spruce,
are common at this elevation. The Krummholtz growth form
is created by blowing ice crystals shearing the needles and
branches from the trees. Other vegetation in this area
includes common juniper (Juniperus communis), cascade
and arctic willows (Salix cascadensis, S. arctica), alpine
fescue (Festuca ovina), and tufted hairgrass.
Along the ridge above French Creek, you can see 50-yearold clearcuts that have not revegetated because of the
blowing snow and extremely harsh growing conditions at
this elevation. This highway closes to travel each winter
beginning with the first heavy snows in September or October and reopens usually by Memorial Day, although it has
sometimes been closed through part of June.
Stop 8: Fen in a Kettle Hole on the Sand Lake Road.
Elevation is 8,600 ft (2,621 m). Present on the fen are bog
birch (Betula glandulosa) and at least four different willows,
including planeleaved willow (Salix planifolia) and
shortfruited willow (S. brachycarpa). On drier ground around
the fen, you’ll find mountain silver sagebrush (Artemisia
cana var. viscidula).
USDA Forest Service Proceedings RMRS-P-31. 2004
Soils and Vegetation of the Snowy Range, Southeastern Wyoming Wildland ...
Munn and Kinter
The soil here is a Cryofibrist. The pH of this organic soil is
controlled by Nash dolomite rocks in the moraines around
the kettle hole. As that rock has weathered, it buffers the pH
of the system, both for the mineral forest soils in the upland
and the organic soil. These soils all have nearly neutral pH.
Normally you would expect fairly acidic soils under the trees
and also in a bog setting like this. But because the Nash
Dolomite occurred near the point of ice collection, blocks of
it were carried here, causing the neutral pH.
Organic soils occur in sites with poor drainage and high
water tables in the Snowy Range. Their use here is limited
by the short growing season. In warmer climates, Histosols
are prized for high value crops such as vegetables because
of their generally excellent properties as a growing media.
Organic soils have high water-holding capacities (200 to
600 percent by weight) and do not have problems with
aluminum toxicity that is common for acid mineral soils.
Organic soils present special management problems if they
are drained: they subside as microbes accelerate decomposition of the organic material in the newly well-aerated soil.
After drainage, they also are prone to wind erosion and
burning. Histosols have extremely low bearing strength
and are unsuitable foundations for houses and roads. The
soil has a very thin Oa horizon (approximately 5 cm) over
perhaps 3 m of Oi horizons, underlain by a 2C (Pinedale
till). The soil is a Fibrist at the suborder level of Soil
Taxonomy because the fibers hold up well when rubbed; the
inference is that they are only slightly decomposed.
where this soil occurs are all on the order of 1,600,000 years
plus. This Bkm horizon material is what the locals refer to
as “caliche” in the Southwest. The laminar layer at the top
of the Bkm indicates that water no longer penetrates it at all.
The original parent material here was quartzite-rich alluvium from the Little Laramie River, and the carbonate
accumulation represents a tremendous accumulation of
dust over time. These petrocalcic horizons are a very effective “armor” against erosion for the old valley floors. The
high surfaces in the southern Laramie Basin are old alluvial
valley floors, which are still extant because of the protection
afforded them by the petrocalcic horizons and alluvial cobble
channel deposits which occur on them. Plants here include
winter fat (Krascheninnikovia lanata), ricegrass (Oryzopsis
hymenoides), and wheatgrasses.
In the bottom of the Libbey Creek drainage we are
driving through very classic Pinedale moraines, short steep
slopes with a lot of stone showing in the road cuts. The ridges
above us to the south have Bull Lake till on them. During the
Pinedale, glaciers cut the deep canyons. The Bull Lake
glacier carried till further out into the basin, and as we pass
through the terminal Pinedale moraine we come to a more
smooth and rounded topography that is the terminal moraine of the Bull Lake glacier. The landscape on the terminal
moraine of the Bull Lake glacier has a more rounded form
and is still studded with boulders of quartzite. Wyoming big
sagebrush grows here on the terminal moraine and around
the slopes on the igneous rock. Once you get out into the
alluvium in the center of the valley, you lose the sagebrush.
Along the road in the Centennial area at 8,000 ft (2,440 m)
elevation, we again see sedimentary rock exposed in hogback ridges. The ridge above Centennial was the terminal
moraine for the Bull Lake glaciation. There is a veneer of
Bull Lake till above Centennial and a large outwash plain
below the town. Sedimentary rock exposed south and west of
the town includes some very red rocks of Jurassic and
Triassic age. These have a very distinctively bright red
chroma. The exposed sedimentary rocks support mountain
mahogany stands (Cercocarpus montanus), along with bitterbrush, willows, and chokecherry. On the east side of the
Centennial Valley there is a similar uplifted banding of
sedimentary rock against the flank of Sheep Mountain.
Stop 9: Petrocalcic horizon (caliche) on high alluvial surface above the Big Hollow. This stop shows the
maximum expression of carbonate accumulation in Wyoming. The soil is Typic Petrocalcid, loamy-skeletal, carbonatic,
frigid. Horizons are Ak, Bk, Bkm, Bk’, and 2C. The surfaces
USDA Forest Service Proceedings RMRS-P-31. 2004
Along the Highway to Laramie: The Big Hollow is
the largest deflation feature in North America, and you can
see the entire rim around the Hollow. Apparently wind was
channeled between Sheep and Jelm Mountains, and scoured
out the hollow during Late Pleistocene time. The rims on
either side are actually old alluvial valley floors, and the
cobbles from the alluvium protected those surfaces on the
north and south from erosion. The area where the basin lies
would have been a small shale hill complex prior to the
deflation. The shale hill in the distance is an example of the
landscape before the topographic reversal. The shale hill
was in place when the alluvial valleys were deposited on
either side, but was much softer and eroded away, and so you
now have a depression. Where the alluvial valley floors ran
from the mountains toward Laramie on either side of that
shale hill, the alluvium protected those surfaces from erosion. Today, they are still intact and stand high above the
Hollow. The small pumping station is the southern end of the
Quealy Dome oilfield. The pumping station reflects current
oil prices; when prices are high (above $24 per barrel), the
well is active, and when prices fall below that mark, the field
is closed down.
The white rock that you can see in the distance is cretaceous Niobrara chalk formation. The Niobrara formation
underlies the alluvial valley floor that we follow all the way
into Laramie. Between here and town, ground water availability is poor, and livestock owners have dug stock ponds.
By placing snow fence around the pond, they can successfully trap blowing snow in wet winters to collect stock water
on these high surfaces for use in the summer.
Return to the University of Wyoming Campus.
References _____________________
Case, J. C.; Arneson, C. S.; Hallberg, L. L. 1998. Preliminary
1:500,000-scale digital surficial geology map of Wyoming: Wyoming State Geological Survey, Geologic Hazards Section Digital
Map 98-1 (HDSM 98-1). Laramie: Wyoming State Geological
Survey.
Hausel, W. D. 1993. Guide to the geology, mining districts, and ghost
towns of the Medicine Bow Mountains and Snowy Range Scenic
Byway. Public Inf. Circ. 32. Laramie: The Geological Survey of
Wyoming. 53 p. plus maps.
Kim, H. B. 1993. Soil morphology and genesis along an elevational
transect, Snowy Range and Laramie Basin, Wyoming. Laramie:
University of Wyoming. Thesis.
Love, J. D.; Christianson, A. C. 1985. Geology of Wyoming. Reston,
VA: U.S. Department of the Interior, U.S. Geologic Survey.
215
Munn and Kinter
Soils and Vegetation of the Snowy Range, Southeastern Wyoming Wildland ...
McCahon, T. J.; Munn, L. C. 1991a. Morphology and genesis of
paleosols in the Medicine Bow Mountains, Wyoming. Soil Science. 152: 472–484.
McCahon, T. J.; Munn, L. C. 1991b. Soils developed in Late Pleistocene till, Medicine Bow Mountains, Wyoming. Soil Science. 152:
377–388.
Mears, B., Jr. 2001. Glacial records in the Medicine Bow Mountains
and Sierra Madre of southern Wyoming and adjacent Colorado,
with a traveler’s guide to their sites. Public Inf. Circ. 41. Laramie:
The Geological Survey of Wyoming. 26 p.
Mitchell, R. A. 1985. Characterization and interpretation of a buried
paleocatena near Jelm Mountain, Wyoming. Laramie: University of Wyoming. Thesis.
Munn, L. C. 1987. Soil genesis associated with periglacial ice wedge
casts, southcentral Wyoming. Soil Science Society of America
Journal. 51: 1000–1004.
Munn, L. C. 1992. Periglacial features as sources of variability in
Wyoming Aridisols. In: Kimble, J. M., ed. Proceedings of the sixth
international correlation meeting (VI:ISCOM) characterization,
classification and utilization of cold Aridisols and Vertisols.
Lincoln, NE: Soil Conservation Service, National Soil Survey
Center: 133–137.
Munn, L. C. 1993. Effects of prairie dogs on soils and associated
plant communities. In: Oldemeyer, J. L.; Biggins, D. E.; Miller, B.
J.; Crete, R., eds. Management of prairie dog complexes for the
reintroduction of the black-footed ferret. Biological Rep. 13.
Washington, DC: U.S. Department of the Interior, Fish and
Wildlife Service: 11–17.
Munn, L. C.; Spackman, L. K. 1991. Soil genesis associated with
periglacial ground wedges, Laramie Basin, Wyoming. Soil Science Society of America Journal. 55: 772.
Nelson, B. E. 1984. Vascular plants of the Medicine Bow Range.
Laramie, WY: Jelm Mountain Press.
Rahman, S.; Munn, L. C.; Vance, G. F.; Arneson, C. 1997. Wyoming
Rocky Mountain forest soils: mapping using an ARC/INFO geographic information system. Soil Science Society of America
Journal. 61: 1730–1737.
Reider, R. G.; Gurley, S. R. 1975. Pre-Wisconsin paleosols and
related soils on the South French Creek-Middle Fork (Little
Laramie River) interfluve, Medicine Bow Mountains, Wyoming.
Contributions to Geology. 14: 41–49.
Reider, R. G.; Kuniansky, N. J.; Stiller, D. M.; Uhl, P. J. 1974.
Preliminary investigation of comparative soil development of
Pleistocene and Holocene geomorphic surfaces of the Laramie
Basin. In: Wilson, M., ed. Applied geology and archeology: the
Holocene history of Wyoming. Geological Survey of Wyoming
Rep. Invest. 10. Laramie: Wyoming Geological Survey: 27–33.
Soil Survey Staff. 1999. Soil taxonomy: a basic system of soil
classification for making and interpreting soil surveys. 2d ed.
USDA-SCS Agric. Handb. 436. Washington, DC: U.S. Government Printing Office.
Spackman, L. K.; Munn, L. C. 1984. Genesis and morphology of soils
associated with formation of Laramie Basin (Mima-like) mounds.
Soil Science Society of America Journal. 48: 1384–1392.
Thybony, S.; Rosenberg, R. G.; Rosenberg, E. M. 2001. The Medicine
Bows, Wyoming’s mountain country. Caldwell, ID: Caxton Press.
180 p.
216
USDA Forest Service Proceedings RMRS-P-31. 2004
You may order additional copies of this publication by sending
your mailing information in label form through one of the following
media. Please specify the publication title and number.
Fort Collins Service Center
Telephone
(970) 498-1392
FAX
(970) 498-1396
E-mail
Web site
Mailing Address
rschneider/@fs.fed.us
http://www.fs.fed.us/rm
Publications Distribution
Rocky Mountain Research Station
240 Prospect Road
Fort Collins, CO 80526
Pesticide Precautionary Statement
This publication reports research involving pesticides. It does not
contain recommendations for their use, nor does it imply that the
uses discussed here have been registered. All uses of pesticides
must be registered by appropriate State and/or Federal agencies
before they can be recommended.
CAUTION: Pesticides can be injurious to humans, domestic
animals, desirable plants, and fish or other wildlife—if they are not
handled or applied properly. Use all pesticides selectively and
carefully. Follow recommended practices for the disposal of
surplus pesticides and pesticide containers.
CAUTION:
PESTICIDES
The use of trade or firm names in this publication is for reader information and does not
imply endorsement by the U.S. Department of Agriculture of any product or service.
Federal Recycling Program
Printed on Recycled Paper
The Rocky Mountain Research Station develops scientific information and technology to improve management, protection, and use of
the forests and rangelands. Research is designed to meet the needs
of National Forest managers, Federal and State agencies, public and
private organizations, academic institutions, industry, and individuals.
Studies accelerate solutions to problems involving ecosystems,
range, forests, water, recreation, fire, resource inventory, land reclamation, community sustainability, forest engineering technology,
multiple use economics, wildlife and fish habitat, and forest insects
and diseases. Studies are conducted cooperatively, and applications
may be found worldwide.
Research Locations
Flagstaff, Arizona
Fort Collins, Colorado*
Boise, Idaho
Moscow, Idaho
Bozeman, Montana
Missoula, Montana
Lincoln, Nebraska
Reno, Nevada
Albuquerque, New Mexico
Rapid City, South Dakota
Logan, Utah
Ogden, Utah
Provo, Utah
Laramie, Wyoming
*Station Headquarters, 240 West Prospect Road, Fort Collins, CO 80526
The U.S. Department of Agriculture (USDA) prohibits discrimination in all its
programs and activities on the basis of race, color, national origin, sex, religion,
age, disability, political beliefs, sexual orientation, or marital or family status. (Not
all prohibited bases apply to all programs.) Persons with disabilities who require
alternative means for communication of program information (Braille, large print,
audiotape, etc.) should contact USDA’s TARGET Center at (202)-720-2600 (voice
and TDD).
To file a complaint of discrimination, write USDA, Director, Office of Civil Rights,
Room 326-W, Whitten Building, 1400 Independence Avenue, SW, Washington, DC
20250-9410 or call (202)-720-5964 (voice or TDD). USDA is an equal opportunity
provider and employer.